Full Text Archive logoFull Text Archive — Free Classic E-books

Scientific American Supplement, No. 492, June 6, 1885 by Various

Part 1 out of 2

Adobe PDF icon
Download this document as a .pdf
File size: 0.2 MB
What's this? light bulb idea Many people prefer to read off-line or to print out text and read from the real printed page. Others want to carry documents around with them on their mobile phones and read while they are on the move. We have created .pdf files of all out documents to accommodate all these groups of people. We recommend that you download .pdfs onto your mobile phone when it is connected to a WiFi connection for reading off-line.

Produced by Don Kretz, Juliet Sutherland, Charles Franks and the DP Team



NEW YORK, JUNE 6, 1885.

Scientific American Supplement. Vol. XIX, No. 492.

Scientific American established 1845

Scientific American Supplement, $5 a year.

Scientific American and Supplement, $7 a year.

* * * * *


I. ENGINEERING AND MECHANICS.--The New Spanish Artillery.--2

Qualitative Tests for Steel Rails.--By L. TETMAJER.

A New Form of Small Bessemer Plant.--By A. TRAPPEN.

Triple Compound Engines.--A paper read by A.E. SEATON
before the Institution of Naval Architects.

Early History of the Steam Jack.

Bridge over the River Adige, at Verona.--13 figures.

Pumping Machinery.--Mine pumps.--Direct acting steam pumps.

Improved Gun Pressure Gauge.--2 figures.

Measuring the Thickness of Boiler Plates.

On an Express Engine.

II. TECHNOLOGY.--Improved Plaiting Machine.--With engraving.

Self-acting Shuttle Guard.--1 figure.

Ruler and Triangle for Hatching.

The Distillation of Sea Water.--1 figure.

Aids to Correct Exposure on Photographic Plates.--An
interesting paper by W. GOODMAN.

Isochromatic Photography.--By FRED. E. IVES.--2 figures.

Distortion from Expansion of the Paper in Photography.

III. ELECTRICITY, ETC.--On the Fritts Selenium Cells and
Batteries.--A paper read before the American Association

Electricity Applied to the Manufacture of Varnish.--2 figures.

Naglo Brothers' Telephone System.--3 figures.

The Gerard Electric Lamp.--1 figure.

A New Reflecting Galvanometer.--3 figures.

IV. ART AND ARCHITECTURE.--Groups of Statuary for the Pediment
of the House of Parliament in Vienna.--2 engravings.

The Casino at Monte Carlo.--An engraving.

V. PHYSICS.--Determining the Density of the Earth.--1 figure.

Physics without Apparatus.--The Porosity and Permeability of
Bodies.--A Hot Air Balloon.--2 engravings.

VI. NATURAL HISTORY.--Winter and the Insects.--An engraving.

Silk Worm Eggs.--With engraving.

VII. HORTICULTURE.--The Melleco.--Ullucus tuberosa.--With

VIII. PHYSIOLOGY, MEDICINE, ETC.--Histological Methods.--Section
cutting machines.--Methods of preserving the
tissues.--Preservative media.--Preparation for mounting
tissues.--1 figure.

Life History of a New Septic Organism.

Erythroxylon Coca as a Therapeutic Agent.--By Dr. E.R. SQUIBB.

* * * * *


The Spanish Government is now engaged in supplying some of its principal
fortifications with heavy guns of the most improved construction. The
defenses of Cadiz and Ceuta have been greatly strengthened in this
respect. The most recent additions are some very powerful Krupp guns for
the fortress of Isabel II., at Mahon.


We give engravings from photographs, as presented in _La Illustracion
Espanola_. These guns are breech loaders, of steel, 301/2 centimeters
caliber, or 12 inches, 49 tons weight.


One of our engravings shows the great revolving crane by which the guns
were lifted and placed on the truck for conveyance over a track to their
intended position. This crane is worked by eight men, and readily lifts
burdens of about 200,000 lb. The other engraving shows the jack frame and
jacks employed to remove the gun from the temporary truck. At a range of
7,000 yards these guns are able to penetrate iron plates of two feet

* * * * *



This memoir is the first of a series upon the unification of nomenclature
and classification of building materials, undertaken by the author at the
request of the Swiss Engineers' and Architects' Union. For its
preparation numerous mechanical tests have been made upon steel rails,
both good and bad, taken from the Swiss railways, while the corresponding
chemical analyses have been made by Dr. Treadwell in the Polytechnic
Laboratory, at Zurich. The results are given for twenty-two examples,
about one-half of which have stood well, while the remainder have either
broken, split, or suffered considerable abrasion in wear; but in many
instances the mechanical test of tensile strength, elongation, and
contraction, and the figures of quality (Wohler's sum and Tetmajer's
coefficient) deduced from these have varied very considerably for the
results obtained in practice.

The best wearing rails, which often give contradictory results with the
tensile test, were comparatively pure manganese steels, low in silicon,
only exceptionally up to 0.2 per cent., but generally below 0.1 per
cent., and with less than 0.1 per cent. of phosphorus and sulphur. On the
other hand, rails with a tendency to break or split are low in carbon,
with variable proportions of manganese, but contain much silicon, 0.3 to
0.9 per cent., and often above 0.1 per cent. of phosphorus. Another
series of experiments upon rails for the Finland lines made by the author
in 1879-80 shows the high quality of manganese steel. These are
essentially highly carburized (0.3-0.4 per cent. carbon) with 0.7 to 1.4
per cent. manganese, and have stood three and a half years' wear without
a single one being broken; while those of silicon steel with 0.106-0.144
per cent. carbon, 0.592-0.828 manganese, and 0.423-0.435 silicon have
failed in many cases, showing a great tendency to split. In both of the
latter instances, however, the figures deduced from tensile tests of both
good and bad specimens were substantially the same.

The causes of the difference between the two kinds of steel the author
attributes to differences in the structure of the ingot due to the agent
used in "chemical consolidation," which may be either manganese or
silicon, which structures are illustrated by photographs of ingot
fractures. When silicon is used there is a tendency to unsoundness about
the exterior of the ingot, which is surrounded by a honeycomb-like
cellular casing of greater or less depth; while with manganese the
vesicular cavities are more or less dispersed through the whole
substance, or concentrated toward the interior of the ingot. Rails made
from the former are, therefore, more likely to contain unsound portions
near the outer wearing surface, and to give unsatisfactory results in
wear, than those from the latter; but as the test pieces are usually cut
from the center of the railhead, the tensile resistance of the interior
may be equal to or surpass that of the superior material. In summing up
his observations the author concludes that the method of tensile testing
is mainly of value in determining the quality of the material, but that
for the finished product properly arranged falling weight tests are
necessary. He also considers that the test pieces should be flat bars of
2.5 to 3.5 centimeters in area, cut as near as possible to the outer
surface of both head and foot of the rail. He reprobates especially the
research for microscopic imperfections (mikrobensuecherei) upon the
fractured surfaces, as an annoyance to the producer, and perfectly
useless to the consumer.--_Stahl und Eisen_, vol. iv., page 608; through
_Proc. Inst. Civ. Eng_.

* * * * *



The success of the Bessemer process when carried out on the small scale
at Avesta in Sweden, as described by Professor Ehrenwerth, and subsequent
experiments of a similar kind made at Pravali, in Carinthia, and
elsewhere, have led the author, who is specially occupied in the building
of Bessemer works, to design a plant suitable for operation upon small
charges. This consists essentially of a converter about 1 meter outside
diameter, and 1.5 meters high, connected by a single trunnion to a
horizontal steel shaft carried by the arm of a hydraulic crane which is
very similar in character to the ladle crane of a large sized converter.
The sweep of the crane is such as to allow the converter to be brought
close up to the tap hole of the blast furnace or cupola, so that the use
of open gutters for the fluid metal may be avoided as much as possible.
The converter is turned on its axis by a screw and worm wheel, which is
manipulated by a workman standing on a platform at the opposite arm of
the crane. The blast is brought in from above by a pipe down the central
pillar of the crane, which is connected with the blast-main by a flexible
tube and packed joint. The outer trunnion bearing is open, so that by
slightly raising and lowering the ram of the crane, the converter may be
left suspended to a weighing machine in front of the furnace, if it is
required to determine the weight of the charge. When the converter is
filled, it is borne by the crane into a convenient position for blowing,
and if the basic method is followed for removing the slag, the converted
metal is cast into ingot moulds, which are manipulated by a small ingot
crane of the ordinary pattern. In the case of small existing
blast-furnaces, which usually have their tap holes near to the ground, it
may be necessary to have a shallow ingot pit (20 to 24 inches deep); but
with cupolas this will not generally be necessary, and the whole of the
operations may be carried on at the ground level. Each crane is intended
to be supplied with two or three converters, so that operations may be
carried on continuously. The weight of charge proposed is 15 cwt., which
should under ordinary conditions give 12 cwt. of ingots. Taking the time
of a single converting operation at half an hour it will be easy to
obtain fifty blows per day, or a production of 30 tons. This may be
easily increased by placing a second converting crane on the other side
of the furnace, for which the same blowing engine will be sufficient, as
the actual blowing time will not exceed twelve minutes. The labor
required for each converter will be about six men per shift.

The blast required has been experimentally determined at 40-50 cubic
meters per minute at 15 lb. pressure. This will be supplied by a single
cylinder engine of 900 millimeters blast, and 786 millimeters steam
piston, diameter 786 millimeters, stroke making fifty revolutions per
minute, which is also to work a Root blower and the accumulator pumps.
Having regard to these very different demands upon the power of the
engine, it will be provided with expansion gear, allowing a considerable
variation in the cut-off. A single boiler of 70 to 75 square meters
heating surface will be sufficient. The accumulator is intended to work
at 300 lb. pressure.

The cost of the plant, including one of each of the following items,
converter, converter truck, blowing engine, accumulator, ingot crane,
centesimal weighing machine, and accumulator pump, is estimated at L2,050
to L2,100; and that of the steam boiler, L325. The buildings may be of
the simplest and cheapest possible character. As the productive power of
such a plant contrasts very favorably with its cost, the author considers
that it may be fairly expected to meet the competition of large works,
especially in the manufacture of a high-class product.--_Stahl und
Eisen_, vol. iv., page 524; through _Proc. Inst. Civ. Eng_.

* * * * *


[Footnote: Paper read before the Institution of Naval Architects, March
27, 1885.]


My attention was first called to the modern triple compound engine by the
published reports of the trial trip of the yacht Isa, and in it I plainly
discerned the germs of a successful new type of engine; but it was not
until I had seen the engines of the screw steamer Aberdeen erected in the
workshops of Messrs. Robert Napier & Sons that I became convinced that it
was the engine of the immediate future. It is, however, due to the
farsightedness and enterprise of Mr. C.H. Wilson, M.P., that I was
enabled to try the merits of the new system and compare it with the old.
Mr. Wilson had already viewed the triple compound engine with more than
ordinary interest, and it required little persuasion on my part to allow
the company to which I have the honor to belong to construct a triple
expansion engine in lieu of the ordinary compound for one of four sister
ships which it then had in hand for Messrs. Thomas Wilson, Sons & Co.,
the latter only stipulating that it was to be of the same power as the
engine already contracted for. As I was quite convinced that economy was
due to the system rather than to the higher pressure, it was decided not
to increase the boiler pressure more than was necessary to suit the
triple system. The other three ships already alluded to were being fitted
with engines having cylinders 25 inches and 50 inches diameter by 45
inches stroke, and supplied with steam of 90 lb. pressure from a double
ended boiler 13 feet 9 inches diameter by 15 feet long, having a total
heating surface of 2,310 feet, so that these engines have every
qualification for being economical so far as general proportions go, the
stroke being an abnormally long one and the boiler of ample size.
Experience has since shown that these engines are economical in coal,
and the wear and tear exceptionally small.

The new engines for the fourth boat were made with considerably shorter
stroke, and the cylinders proportioned so as to give equal power; they
are 21 inches, 32 inches, and 56 inches diameter by 36 inches stroke, the
high pressure cylinder being supported on columns immediately over the
medium cylinder, and in other respects these engines were made as near as
possible like the other ones above named. Steam at 110 lb. pressure is
supplied from a double ended boiler 12 feet 9 inches diameter and fifteen
feet long, having a total heating surface of 2,270 square feet, and
identical in design with the boiler supplied for the other engines. The
propellers were made exactly alike in all respects, and the ships being
likewise precisely alike, a comparison of the performances of the one
fitted with the triple engines could be made with as little grounds for
differences of opinion as is possible. One of the ships fitted with the
ordinary compound engines was named the Kovno, that with the triple
compound engines the Draco. Their dimensions are as follows:

Feet. Inches.
Length between perpendiculars. 270 0
Breadth. 34 0
Depth of hold. 18 3

And of 1,700 tons gross register. They are ordinary cargo boats, built of
steel, having a raised quarter deck and long bridge amidships, but
nothing about them otherwise requires comment.

After making a voyage or two to the Baltic, and finding that everything
was working satisfactorily, the Kovno was loaded with 2,400 tons dead
weight, and sailed in January, 1883, for Buenos Ayres; the Draco was
loaded with 2,425 tons dead weight, and sailed March, 1883, for Bombay,
the distance in both cases being about 6,400 miles. It was thought
advisable, for purposes of comparison, that the ships should steam at as
near as possible the same speed; and to attain this object, we considered
the safest plan was to instruct the engineers as to the average amount of
coal they were to burn per day, and experience with these ships on their
Baltic voyages had fixed this at 12 tons in the case of the Kovno and 10
tons in the case of the Draco. During the voyage each ship seems to have
had fair average weather, and equal care was taken in getting the best
results possible. The average speed of the Draco was, however, 8.625
knots, or 207 miles per day, the engines making on the average 57.5
revolutions per minute, while the Kovno did only 8.1 knots, or 194 miles
per day, the engines making 55.5 revolutions. The coal used was ordinary
South Yorkshire, just as it comes from the pits for bunker purposes. The
indicated horse power in each case would average about 600. The total
coal consumed was 326 tons in the Draco and 405 tons in the Kovno, or a
saving of 19.5 per cent. over the ordinary compounds, with an increase of
speed of 6.5 per cent.

In December, 1883, one of the others, the Grodno, sailed from Bombay, and
attained an average speed of 8.5 knots, or 204 miles per day, the engines
making 57 revolutions, with a coal consumption of 12.8 tons per day, or
469 tons on the voyage. The Draco's consumption is therefore 30.5 per
cent. less than that of the Grodno on the round voyage, and 20.3 percent
per day.

The success of the triple compound engine was in these instances more
than had been anticipated, and induced Mr. Wilson to go a step further.
The S.S. Yeddo had been refitted with boilers made for a working pressure
of 90 lb. per square inch, but owing to the size of the shafting the
working pressure was limited to 70 lb.; the average consumption of coal
under these circumstances on two voyages was 17 tons per day. These
boilers had a margin of safety beyond what was required by the rules when
made, and as the Board of Trade rules had been modified in the mean
while, it was found that they could with safety be worked at 100 lb. per
square inch. A third cylinder was now fitted on the top of the original
low pressure, and the safety valves loaded to the 100 lb., and the ship
was dispatched to Cronstadt. After making two voyages under similar
circumstances to the two previous ones, the average consumption was 13.5
tons per day only. In this case it was the same ship, same boilers, same
engines, same propeller, and same men, the only difference being the
addition of a third cylinder and the increase of pressure.

So far all the trials had been made with two crank engines; so it was now
decided to construct another set of engines for 150 lb. pressure, having
a crank to each cylinder. These engines had cylinders 201/2 inches, 33
inches, and 58 inches diameter by 36 inches stroke, and were fitted into
the screw steamer Rosario, whose dimensions are 275 feet 3 inches between
perpendiculars, 34 feet 3 inches beam, and 19 feet 2 inches depth of
hold, 1,862 tons gross, and the deadweight capacity 2,550 tons. In March
last year she was loaded with 2,530 tons deadweight, and did the voyage
to Bombay at an average speed of 8.6 knots on a consumption of 10.5 tons
per day of South Yorkshire coal, and burnt on the voyage 347 tons. This
result is superior to that of the Draco when the size of the ship is
taken into account, but is not so much so as might have been anticipated
from the increase of pressure and the rate of expansion, which was 14.4
in the Rosario and 12 in the Draco. Another set of engines was made from
the patterns of those of the Draco, but with the high pressure cylinder
20 inches diameter, steam at 150 lb. pressure being supplied from two
single ended boilers, having a total heating surface of 2,200 square
feet. They are fitted in the S.S. Finland, a cargo boat 270 feet long, 35
feet beam, by 18 feet depth of hold, and 1,954 tons gross register. In
January she was loaded with 2,500 tons deadweight, and sailed for
Rangoon. The average speed attained was 8.42 knots per hour, or 202 miles
per day, on a consumption of 10.3 tons of Welsh coal per day, the rate of
expansion being 12. It should be mentioned that all these ships named are
fitted and steered with steam stearing gear, so that in comparing these
results and those published of the engines made by an eminent engineer in
the north of England, an allowance should be made, as in that ship there
was no steam stearing gear.

I have chosen to make all these comparisons by reference to the ships'
logs, and to give results such as a shipowner looks for rather than those
which engineers prefer to use in forming a judgment on the merits of
different engines. I do this for two reasons: first, because the
commercial success of the triple compound engine depends on the saving it
can effect in a long voyage; and secondly, because I had no reliable
indicator diagrams from which the consumption per indicated horse power
could be calculated with any degree of accuracy. On trial trips with the
steamers already named, the consumption of ordinary South Yorkshire coal
was 1.6 lb. per indicated horse power, and the consumption of water per
indicated horse power calculated from the high pressure indicator
diagrams was 1.41 in the Draco, 13.2 in the Rosario, and 13.16 with the
Finland, or taking the medium pressure diagrams, it was 12.2, 1.30, and
11.95 respectively. Twelve months ago we constructed for Messrs. Thomas
Wilson, Sons & Co., two sets of triple expansion engines of 600 indicated
horse power, one having two cranks and the other three cranks, the
engines, boilers, and propellers being otherwise exactly alike and fitted
into sister ships. The water consumed in the three crank engine is 12.93
lb., against 13.0 in the two crank, but the former drives its ship nearly
1/2 knot per hour faster than the latter does its, and when both ships are
driven at the same speed the consumption of coal in the three crank ship
is considerably less than in the other.

We have now entirely given up the construction of two-crank triple
expansion engines, because of the impossibility of equally dividing the
work between the cranks; for, although the engine when running appeared
to be perfectly balanced, the wear of the brasses of the crank having the
two cylinders was always considerably more than that of the other.
Placing the high pressure cylinder over the low pressure cylinder seemed
to give the most satisfactory results, but even these were far inferior
to those once obtained with the three cranks. We have lately constructed
some very small three-crank engines from which exceedingly good results
were obtained; the cylinders are only 111/2 inches, 17 inches, and 30
inches by 18 inches stroke, which developed 218 indicated horse power
with a consumption of 12.8 lb. of water per indicated horse power, and
this, together with some other observations, leads me to believe that the
best economical results will be obtained by running triple expansion
engines at a much higher number of revolutions than is usual, and with a
rate of expansion not less than 12 for a steam pressure not less than 140
lb. (155 absolute). The largest engines we have made of this type so far
are those of S.S. Martello, which have cylinders 31 inches, 50 inches,
and 82 inches diameter by 57 inches strokes and indicate at sea 2,400
horse power when running at 60 revolutions with steam of 150 lb.
pressure; the consumption of Yorkshire coal is 37 tons per day average
throughout a New York voyage. Had Welsh coal been used in every case, the
results would have been very much better, for, in addition to the
superior evaporative power of Welsh coal, it is slow burning and much
more easily controlled, especially on the comparatively short grates of
these modern boilers, the quick-burning Yorkshire coal causing the safety
valves to frequently blow off when working near the load pressure unless
great care is taken by the firemen.

I trust these few particulars may be of interest to the Institution, and
especially to those members of it who are particularly interested in the
commercial success of our mercantile navy. I have purposely avoided
engineering details and technicalities of any kind, giving only such
information as will tend to give British shipowners faith in that form of
engine which will undoubtedly help them to successfully tide over bad
times, and keep the bulk of the carrying trade of the world in their

* * * * *


_To the Editor of the Scientific American:_

A friend has brought me a copy of the SCIENTIFIC AMERICAN SUPPLEMENT, of
April 18, 1885, containing an article about a "steam jack."

Says Mr. J.G. Briggs, in the _American Engineer:_ "Of its origin nothing
is known." Also the invention is attributed to "Benjamin Baleh." I can
give you the true history of the "steam jack." It was invented by my
grandfather, John Bailey, of Hanover, Plymouth County, Mass. He was a
minister of some note in the Society of Friends, or Quakers.--a man of
superior mental ability, but poor in purse, for, like all early
inventors, he reaped but little pecuniary benefit from his inventions.
Among those inventions was the first iron sink in this country--if not in
the world. A few years ago that sink was in use at his old home in
Hanover. He also invented the crooked nose for the tea-kettle. Previous
to that the nose was straight. Both sink and tea-kettle were cast at the
Middleborough foundry. When he made the steam-jack he said, "In less than
fifty years the common mode of travel would be by steam." People called
him "steam mad." But about the jack. We have one in our possession of
which your cut is an exact copy. We have used it several times. We also
have the parchment _patent_, of which I send you a copy. The jacks were
not in general use, for soon after the invention the "tin kitchen," or
"Dutch oven," as it was sometimes called, was introduced, and superseded
the jack entirely, as people were afraid of being blown up by steam. The
patent says, "John Bailey, of Boston," showing that at that early date
Boston was considered the _Hub_, and that it was considered a good thing
to hail from there. Hanover is about twenty-four miles from Boston.

Trusting I have not wearied you, I am,


Bleak House, Lynn, Mass., May 12.


_United States_.

To all to whom these Presents shall come, Greeting. Whereas, John Bailey,
of Boston, in the State of Massachusetts, hath presented a petition to
the Secretary of State, the Secretary for the Department of War, and the
Attorney-General of the United States, alledging and suggesting that he
hath invented the following useful Machine, not before known or used,
that is to say: A Steam Jack, consisting of a boiler, three wheels, and
two wallowers; the steam which issues from boiling water in the said
boiler gives motion to one of those wheels by striking on buckets on its
circumference; on the outer end of the axle of the wheel is a wallower,
the rounds of which fall into the teeth of a second wheel; on the axle of
this second wheel is another wallower, the rounds of which fall into the
teeth of a third wheel; on the axle of which third wheel is a spit: and
praying that a patent may be granted therefor: and, whereas, the said
invention hath been deemed sufficiently useful and important: These are,
therefore, in pursuance of the Act, intitled an Act to promote the
progress of useful arts, to grant the said John Bailey, his heirs,
administrators, or assigns, for the term of fourteen years, the sole and
exclusive right and liberty of constructing, using, and vending to others
to be used, the said invention so far as he the said John Bailey was the
inventor, according to the allegations and suggestions of the said
petition. In Testimony whereof I have caused these Letters to be made
patent, and the Seal of the United States to be hereunto affixed. Given
under my hand, at the City of Philadelphia, this twenty-third day of
February, in the year of our Lord one thousand seven hundred and
ninety-two, and of the Independence of the United States of America the
Sixteenth. Go. WASHINGTON.

By the President,


CITY OF PHILADELPHIA, February 23, 1792.

I do hereby certify that the foregoing Letters-patent were delivered to
me in pursuance of the Act intitled an Act to promote the progress of
useful arts: that I have examined the same, and find them conformable to
the said Act.


_Attorney-General of the U.S._


* * * * *



The iron bridge which spans the Adige at Verona, of which we publish
illustrations, has been recently completed to replace an old masonry
bridge built in the fourteenth century, and which was destroyed by the
celebrated flood of 1882. In designing the new work two leading
conditions had to be fulfilled, namely, that there should be a single
opening of 291 ft. between abutments, and that this width should be left
quite unobstructed, for the river is subject to floods, which are
frequent, and very violent and sudden. For this latter reason an ordinary
form of arch, with the roadway above it, was inadmissible, since the
waterway would be seriously obstructed; the special form illustrated was,
therefore, carried into execution. The bridge, as will be seen from Figs.
1, 2, 3, and 7, consists of two main arched girders, with two vertical
sides in lattice work; these arches spring below the level of the roadway
and rise to a considerable height above it, in the center. The horizontal
girders carrying the roadway, are connected to the arches by verticals of
the form and section shown in the drawings. The longitudinal girders are
of double trellis, as will be seen by reference to Figs. 1, 12, and 16.
The following are the principal dimensions of the bridge:

Ft. In.
Clear opening between abutments 291 4
Rise of arch 32 93/4
Width of bridge 37 43/4
Depth of arched girders 4 7


The arched girders are connected together, in the central portion, by a
system of diagonal bracing, as is shown on Figs. 2 and 7. The carriage
road on the platform consists of buckled plates resting on transverse
girders spaced 6 ft. 6 in. apart, and covered with road metal, and for
the sidewalks checkered plates are used. The ironwork in the bridge
weighs 400 tons, and cost 8,400 _l._; the abutments cost 3,600_l._,
making the total outlay on the structure 12,000_l_. The bridge was
tested by a uniformly distributed load of 82 lb. per sq. ft., and under
this stress the arched girders deflected 1.06 in. The horizontal and
vertical oscillation of the bridge, which were carefully observed and
graphically recorded by special instruments, were very slight. The
engineer of the work was Mr. G.B. Biadego, of Genoa.--_Engineering_.

* * * * *


[Footnote: A paper read at the Montreal Meeting of the British

By E.D. LEAVITT, JR., Cambridgeport, Mass.


One of the earliest steam engines, of any size, introduced into America,
was erected about the year 1763, at the Schuylkill copper mine, situated
upon the Passaic River, in New Jersey. All its principal parts were
imported from England; and a Mr. Hornblower (the son, it is believed, of
the well known engineer of that name) came to this country for the
purpose of putting up and running this engine.

At the time when the manufacture of the engines for the Philadelphia
Water Works was commenced, and as late as the year 1803, we find five
engines, in addition to the one above mentioned, noticed as being used in
this country: two at the Philadelphia Water Works; one just about being
started at the Manhattan Water Works, New York; one in Boston; and one in
Roosevelt's sawmill, New York; also a small one used by Oliver Evans to
grind plaster of Paris, in Philadelphia. Thus, at the period spoken of,
out of seven steam engines known to be in America, four were pumping

In the coal regions of Pennsylvania, a simple, high pressure, single
acting Bull engine has been extensively adopted; the dimensions usually
run from 36 inches to 80 inches in diameter, and a very common stroke is
10 feet. At the Empire shaft, in the Schuylkill coal region, there is a
very fine pair of these engines, with 80 inch cylinders, working 24 inch
pumps. The stroke of both steam pistons and pumps is 10 feet. These Bull
engines are placed either vertically or on an incline, as is most
convenient for the workings. The water valves are made either double,
triple, or four beat, according as the pumps are large or small; and the
beats are usually flat, and faced with leather. Many flap-valves are also
in use. These are frequently arranged on conical seats, and work very

The Bull engines, from their strength and simplicity, give very little
trouble, working year after year with astonishing freedom from accident
and slight cost of repair. No attempt is made to economize fuel, which
consists mainly of culm, which would otherwise be wasted. Of late, direct
acting steam pumps placed under ground have found much favor with mine
operators, on account of their portability and small first cost. They
usually range in size from 8 inch steam and 5 inch water cylinders by 12
inch stroke to 80 inch stream and 14 inch water cylinders by 36 inch
stroke. Great numbers of these pumps are in use all over the United

A pumping engine that is remarkable for its size and peculiarities of
construction is located at the Lehigh zinc mine, at Friedensburg, Pa. It
was designed by Mr. John West, the company's engineer, and built by
Merrick & Sons, of the Southwark Foundry, Philadelphia. It is a beam and
fly-wheel engine, the steam cylinder being 110 inches in diameter, with a
stroke of 10 feet. There are two beams on the same main center, from the
outer end of which a double line of bucket and plunger pumps is operated.
The crank-shaft is underneath the steam cylinder; and there are two
fly-wheels, one on each end of said shaft, the crank-pins being fast in
the hubs of the same. There are two connecting rods, which are attached
one to each end of an end beam pin 28 inches in diameter. The main center
and crank shafts are also 28 inches in diameter; each of the two plunger
holes is 24 inches by 30 inches in section; and all the working parts are
in proportion to those heretofore mentioned.

Perhaps no mining district has ever had to contend against greater
difficulties in pumping than have faced the engineers of the celebrated
Comstock lode, Virginia City, Nev. The mines are of great depth, in some
instances 3,300 feet; and the water is hot, rising to 160 degrees Fahr.
The machinery collected at this location is of great variety and
magnitude. There are many Davey engines, both horizontal and vertical.
The Union and Yellow Jacket shafts have compound fly wheel engines of
very great power; the former having a beam, and the latter being
horizontal, with cylinders placed side by side, and pistons connected to
a massive cross-head, from the ends of which connecting rods lead to
crank pins located in the hubs of the fly-wheels, which are overhung upon
the ends of the main shaft. From the center of the cross head, a link
runs to the main pump-bob, which operates a double line of 16 inch pumps,
10 foot stroke. The steam stroke is 12 feet. Depth of shaft, 3,300 feet.

The pumping machinery used in the iron and copper districts of Michigan
usually consists of Cornish plunger pumps, which are operated by geared
engines; the latter making from three to sixteen strokes to one of the

The largest plant of this type yet erected is that of the Calumet and
Hecla copper mine, at Calumet, Mich. There are two lines of pumps,
varying in diameter from 7 inches to 14 inches, and with an adjustable
stroke varying from 3 feet to 9 feet. The object of the adjustable stroke
is to diminish the capacity of the pumps in the dry season. Each line of
pumps is driven from a crank placed on a steel spur-wheel shaft 15 inches
in diameter, making ten revolutions per minute. The mortise spur-wheels
have a diameter of 221/2 feet at the pitch line, with two rows of teeth,
each 15 inches face. The pitch is 4.72 inches. Engaging with the mortise
wheels are pinions of gun iron 4 feet 6 inches in diameter, placed on
steel shafts 12 inches in diameter, and making 50 revolutions per minute.
The 12 inch pinion shafts are driven through mortise wheels 12 feet in
diameter, and 24 inches face, by pinions 3 feet 9 inches diameter, which
make 160 revolutions a minute. The pinion shafts are driven through a
wire rope transmission from an engine located 500 feet distant. The rope
wheels are 15 feet in diameter, and make 160 revolutions a minute. The
engine is 4,700 horse power, and, in addition to driving the pumping
machinery, does the hoisting and air compressing for the Calumet mine.

In the same building with the mine pump gearing is a duplicate
arrangement for operating the man engine. In order to operate the mine
pumps and man engine for the Hecla mine, it was necessary to use rock
shafts, which are made of gun iron, and hollow; they are 32 inches in
diameter outside, with 41/2 inches thickness of metal. The pump rock shaft
is 39 feet 41/2 inches long over all, in two sections, and weighs 40 tons.
There are rockers placed on each end of this shaft, one of which is
connected with a crank on the mortise wheel shaft, and the other with the
surface rods that work the pump-bobs. These rods are of Norway pine, 12
inches by 12 inches in section, and 1,000 feet long. There are two bobs,
one above the other, with axes at right angles, each weighing about 25
tons. The connection from the upper bob to the lower has hemispherical
pins and brasses to accommodate vibrations in right angled planes. The
slope of the main pump is 39 degrees, and the machinery has been designed
to raise water from 4,000 feet depth. The pumps are of the usual Cornish
plunger type, with flap valves. There is an auxiliary engine, of the
Porter-Allen type, for driving the pumps and man engines when the main
engine is not working. It makes a 160 revolutions per minute, the same as
the rope wheels The seeming complication of the arrangement is due to the
fact that it had to be adapted to existing works, for increased depths,
and put in without interfering with the daily operation of the mine.

The Calumet & Hecla Mining Company has also an extensive pumping plant at
its stamp mills, which are located on the shore of Torch Lake, about four
and a half miles from the mine. There are located here 3 pumping engines;
two of which have a capacity of 20,000,000 gallons a day, and a third
10,000,000 gallons a day. The water is elevated between 50 and 60 feet,
and is used for treating the stamped rock. Two of the engines are of the
inverted compound beam and fly-wheel type; and the third is a geared
pump, which has a horizontal double acting plunger, 36 inches in
diameter, by six foot stroke, driven from the crank of a spur-wheel

The spur wheel is 12 feet diameter, 24 inches face, and contains 96
teeth. The pinion engaging with it has 27 teeth, and is fast on the
fly-wheel shaft of a Brown horizontal engine, having a cylinder 18 inches
in diameter, and a stroke of four feet. The steam pressure used is 110
pounds per square inch; and the engine has a Buckley condenser. The pump
valves are annular, of brass, faced with rubber, and close by brass
spiral spiral springs. Their external diameter is six inches, and the
lift is confined to 1/2 inch. There are 91 suction and 91 delivery valves
at each end of the pump. The maximum speed of this pump is twenty-six
double strokes a minute.

The largest of the compound engines is named Ontario, and has a vertical
low pressure cylinder 36 inches in diameter, and an inclined high
pressure cylinder 171/2 inches in diameter; the stroke of both being five
feet. These are inverted over a beam, or rocker; and the pistons are
connected to opposite ends of the same.

The beam attachment of the main connecting rod is made to a pin located
above and midway between the pins for piston connections.

The main center of the beam and the crank shaft have their pedestals in
the same horizontal plane. The throw of the crank is five feet. There are
two differential plunger pumps, having upper plungers 20 inches in
diameter, and lower plungers 33 inches in diameter, with a stroke of 5
feet. These pumps are vertical, and placed beneath the engine bed-plate,
to which they are attached by strong brackets. The pump under the low
pressure cylinder is worked directly from its cross-head by an extension
of the piston rod. The other pump is worked by a trunk connection from
the opposite end of the beam. The radius of the beam is but fifty inches,
but the connections to it are made very long by links.

The lower plungers work through sleeves in diaphragms located in the
center of the pumps. In these diaphragms, the openings for the delivery
valves are made. These valves are similar in construction to those
previously described for the horizontal plunger pump. Their diameter,
however, is but 51/4 inches, instead of 6 inches, and there are 72 suction
and 72 delivery valves for each pump. It will readily be seen that the
action of these pumps is similar to that of the bucket and plunger; each
pump having one suction and two deliveries for each revolution of the
engine. The Ontario is designed to run at a maximum speed of 33
revolutions a minute; and the service required of it is to run regularly
144 hours a week, without a stop, which is performed with the utmost

The differential pump was invented and patented, many years since, by a
party named James Ramsden, in Pennsylvania, who designed it for an
ordinary house pump. It was subsequently reinvented by the writer, who
first ascertained that he was not the original inventor upon applying for
a patent. A pump of this description was run at the Hecla mine for
several years, at a speed of 500 feet a minute; and its performance was
in every way satisfactory.


This class of machinery deserves a prominent place, as the number in use
vastly exceeds those of all other types combined.

The first consideration will be given to the Worthington, which is the
pioneer of its type, having been invented by the late Henry R.
Worthington, and patented in 1844. Mr. Worthington's first pump was
designed for feeding boilers. His first water works engine was built for
the city of Savannah, Ga., and erected in 1854. The second engine, which
was the duplicate of the Savannah engine, was erected at the city of
Cambridge, Mass., in the year 1856, and was guaranteed to deliver 300,000
gallons in twenty-four hours to an altitude of 100 feet. It had a high
pressure cylinder 12 inches in diameter, placed within a low pressure
cylinder 25 inches in diameter; the low pressure piston being annular.
The double acting water plunger was 14 inches in diameter, and worked
directly from the high pressure piston rod; the stroke of pistons and
plunger being 25 inches. This engine was tested in 1860, with the result
of a duty equal to 70,463,750 foot pounds per 100 pounds of coal.
Subsequently, a test made by Mr. Frederick Graff, of Philadelphia (long
prominently connected with the Philadelphia Water Department), and the
late Erastus W. Smith, of New York, developed a duty of 71,278,486 foot
pounds per 100 pounds of coal, which long remained the best record in the
United States. In 1863, Mr. Worthington brought out at Charleston, Mass.,
his crowning success, the duplex engine, which fairly deserves to be
placed first among the hydraulic inventions of this century. This engine
has since been more extensively duplicated for water works purposes than
any other, with the possible exception of the Cornish.

* * * * *


The following description of the construction and mode of action is by
Thomas Shaw, M.E., Philadelphia, the inventor.


Fig. 1 represents the gauge secured to small ordnance, the gun shown in
cross section. Fig. 2 represents face view of the gauge and indicator,
exposing a vertical section through the hydraulic portion of the gauge,
on line 3 and 4 of Fig. 1. The same principles of reduction of high
pressure are used in this gauge as in Shaw's hydraulic gauge. It will be
observed that a solid steel piston, E, in the cylinder, A, is provided
with a plunger on its under side, which comes in contact with an elastic
packing, D; the plunger may stand as 1 to A 1,000, or as 1 to A 100, in
point of area of exposed surface, as compared with the large piston head,
as desired. Assuming the proportions to be 1 to A 1,000, the 1,000 lb.
pressure on the plunger means only 1 lb. pressure in the fluid chamber,
above piston head, E, and this greatly reduced pressure is now
susceptible of measurement by any of the ordinary light pressure
instruments for measuring pressures. All the passage ways connecting to
dial gauge, R, with the fluid chamber above piston, E, are filled solid
with fluid, permitting no air spaces that can be avoided. The steel plug,
L, that forms a passage way between the fluid chamber and the dial gauge,
is provided on one side with a small screw hydraulic pump, with a
reservoir supply of fluid. This part is shown in longitudinal section;
the steel plunger, I, is firmly secured to wheel, F, the long hub, H, of
which is provided with a screw thread on its inner side, which thread
screws upon the exterior of pump barrel, K. After first filling the
interior of the pump barrel with fluid, the said hub is screwed upon the
pump barrel, causing the plunger, I, to force the fluid into the fluid
chamber and passage way leading to the dial gauge, causing the hand or
pointer to move to any predetermined pressure on dial, in advance of
pressure applied in the high pressure chamber at D. The purpose
accomplished in this act is to give the least possible movement of the
pointer to record any maximum pressure, as, for example, assuming that
20,000 lb. was the expected pressure from any one explosive, then the
pointer, by the means above described, can be set at, say, 18,000 lb., in
which event the pointer is reduced to the minimum movement of only 2,000
lb. to register 20,000 lb.

It will be evident that much greater accuracy of measurement of maximum
pressures can be obtained by the minimum movement of the pointer, as both
the inertia and the momentum are reduced to the minimum quantity. The
subsidence of pressure resulting from explosives being about as sudden as
the creation of pressure, causes the pointer to move too rapidly for
correct ocular observation, on which account a static electric current is
employed, causing a stream of electric sparks to shoot off from the end
of the pointer, B, to the brass outer ring, M. The gauge is insulated for
that purpose by glass plate, S, which is secured concentrically to the
gauge proper and the ring, M. Binding posts for the electric wires are
provided at O and P, which wires are shown in Fig. 2. A spring clamp, N,
Fig. 2, enables the insertion of chemically prepared or other paper,
which lies against the inner side of brass rim, M, and held in place by
the clamp, N. The electric sparks above spoken of pierce the strip of
paper with small holes and colored marks. These holes, etc, show the
exact limits to which the pointer has traveled under pressure, and thus
an indelible record is kept by the electrical indications shown upon the
strip of paper. The paper can have the pressures corresponding to gauge
printed upon the same, when the holes are made prominent by holding the
paper to the light, exposing an exact indication of the pressures or
explosives operated with.

The gases resulting from the explosives are injurious to the gauge
packings, etc., on which account the bore in gun, W, and the connecting
steel plug, B, are filled with fluid. A screw plug, U, enables the
insertion of the fluid, after first pushing an elastic wad of rubber, B,
or cork, in the bore near the inner wall of the gun, which wad will
prevent the escape of the fluid to the interior, and be sufficiently free
to prevent any interference with the pressures. The patentee and
manufacturer of this gauge is prepared to fill orders up to 50,000 lb.
per square inch. This gauge is made of the best steel, and is very
compact, the weight being inside of twenty-five pounds.

The inventor has heretofore made mercury column gauges for gunpowder
pressures, which were too large for direct attachment to guns, but were
connected with special powder chambers to test the pressure, etc., of
confined explosives. The experience thus gained enabled the construction
of the instrument here shown, which is adapted to direct attachment to
the gun, making it as easy now to measure gunpowder pressures as it had
been, heretofore, to measure steam pressures. The effect of this movement
is to reduce the exaggerated statement of high pressures, obtained from
ordinary sporting powders; these have been accredited with pressures up
to 40,000 lb. per square inch, but they only really gave 22,000 lb. by
actual gauge measurement. Artillerists and ordnance officers have, in
this instrument, a true pulse of the internal pressures of the gun, of
inestimable value when determining the quantity of powder and the proper
weight of shot. These are important matters in ordnance practice.

This gauge is a compact machine, designed to measure and indicate the
quick pressures resulting from gunpowder explosives and the slow
pressures of hydraulic force; the same mechanism used in both cases
permits the ready testing and examination of gauge under hydraulic
pressure, to determine its accuracy, for the more sudden pressure
occasioned by the use of gunpowder.

* * * * *


The principal object the inventors of the machine we illustrate herewith
had in view in designing it was to arrange a mode of working the grip
motion positively, so that the cloth shall be received freely and without
strain or friction before or up to the very instant at which each fold is
completed, and shall then be seized and firmly held. In existing machines
there is not we believe, any arrangement for the accomplishment of this
purpose; it is true, the table upon which the cloth is folded is relieved
at the termination of the stroke of the plaiting knife, but the upper
gripper bar, against which the folds of cloth are pressed upon the return
of the table to its normal position, is stationary, being rigidly fixed
to the sides of the machine. One result of this rigidity is that the
cloth has to be forcibly thrust by the plaiting knife under the upper
gripper bar, and in consequence of the violence involved the fold just
made at the opposite end is dragged out from the grip, making a short
fold, and further, in the case of delicate finishes, giving rise to
damaged goods. Another result of this arrangement, when the cloth is not
pressed against the upper bar, is that it returns with the return stroke
of the plaiting knife, the grip not being made until the knife is clear
of the upper bar; thus the plaits or folds are made of irregular length.


To remedy this and to prevent its occurrence, Messrs. A. Edmeston and
Sons, Manchester, in the plaiting machines they are now manufacturing
make the upper gripper bar movable as well as the table below. Referring
to the illustration, the upper gripper bars, A A, are capable of moving
about the center pins, B B, and when the machine is working are operated
in the following manner:

Upon the shaft, C, which revolves in unison with the crank shaft working
the plaiting levers and knife, are placed two cams, D, one at each end,
inside the main frames. These cams engage with and work two escapement
levers or pallets, E E, upon which rest the feet of four rods, attached
one end to each of the upper gripper bars. Upon these four rods are
helical springs of sufficient strength to hold down, by means of the
grippers to which they are connected, the folds of cloth that have just
been made. The cam, D, is so shaped that when the advancing plaiting
knife and cloth reach the front edge of the gripper bar, the gripper is
raised from the table to admit them freely. The instant the end of the
stroke is reached the anchor pallet or lever, E, escapes from the cam,
and the gripper bar is suddenly forced on to the knife and cloth by the
springs before mentioned, securely retaining the piece in its position.
Simultaneously with the first of these motions the plaiting table itself
is lowered, and, when the plaiting knife reaches the end of its stroke,
is released by means of the levers and chains, F F, which are in
connection with the escapement pallets, E, and partake of their every
motion. These chains are so attached that they exert no effort upon the
table until the escapement lever is moved, thus permitting the plaiting
table to press upward against either one or both of the gripper bars with
the full force imparted to it by the weights and levers, G' G'. The
chains, furthermore, are also threaded over pulleys in such a manner that
they adjust themselves automatically to every position of the table and
to the different thicknesses which the folded cloth acquires.

It will be obvious from this description that in plaiting there is no
more strain put upon the cloth in placing it under the grip than is
necessary to draw it over the table from the feed rollers. This feature
insures perfect immunity from the dragging out of grip, as already
described, and renders the machine very useful for finishers and
makers-up, as the delicacy with which the cloth is handled prevents any
damage being done to the finish of the lightest fabrics. Double cloth
can, of course, be plaited by it equally well, and the precision and
uniformity with which the cloth is plaited makes the machine thoroughly
reliable as a cloth measurer.--_Tex. Manfr._

* * * * *



The annexed illustration shows the essential parts of Hahlo and
Liebreich's improvement, the loom being now at work. The handrail,
shuttle race, and starting handle can be at once recognized, and the
shuttle guard will be seen in its proper position, which position it
rigidly retains as long as the loom is working, but on a stoppage the rod
swings back close underneath the handrail, and quite clear of the reed.
The mode in which this is accomplished we will endeavor to make clear.
The guard is connected to the starting lever by the arrangement shown,
consisting of a stud on the handle, on which, with the movement of the
slay, lever, a, slides. This lever, by means of another lever and a link,
is attached to the shuttle guard by the crank, b, which, by means of the
set screw in the boss, permits the shuttle guard to be adjusted in the
most convenient place. It will be observed that whenever the loom stops
working, whether it is stopped by hand or automatically, the hand lever
has to be moved, and this movement is communicated to the shuttle guard
by the mechanism just described, placing the guard rod beneath the hand
rail, and leaving the whole of the shuttle race free and unencumbered.
The act of starting the loom brings the guard again to the working
position without any extra act having to be performed by the weaver. The
action is entirely automatic, and the weaver has not anything to do that
she has not to do with the present unguarded looms. The arrangement
appeared to ourselves to be a very efficient one, and it has the merit
that the length of the guard can be made greater than the width of the
cloth, a further advantage that will be recognized by practical weavers.

* * * * *


The instrument shown in the cut is the invention of Mr. Maginnis, and is
designed for producing equidistant hatchings. It consists of a short
ruler, A, and a triangle, B, supposed to be one of 45 deg., but which may be
of any angle. The triangle carries two stops, c c, while the ruler is
provided with a conical piece, D, which is slotted, and is held by a
screw. The play that occurs between this conical slide and the stops
varies according to the position of the former.


The apparatus operates as follows: In the figure, the stop to the right
being in contact with the piece, D, a line is drawn along the right side
of the triangle. Then the ruler is made to slide along the triangle until
D touches the other stop, and then the triangle is slid along the ruler
until the stop to the right touches D again. In this position another
line is drawn, and so on. The position of the piece, D, between the stops
is regulated according to the fineness of the hatching to be
done.--_Chronique Industrielle_.

* * * * *


The supplying of the troops at Suakim and in the Soudan with water is one
of the most important items in the whole conduct of the Egyptian war.
Even in cold or temperate latitudes fresh water is a first necessity for
animal life; much more is this the case in the desert; and the wells in
the country forming the scene of our military operations form in
themselves valuable strategical points. Their supply, however, has to be
supplemented, and to do so artificial means and the aid of the engineer
have to be enlisted into this service.

Many of our readers see notices from time to time in the newspapers about
this or that ship being employed, or at least her steam fittings, in
distilling water for the use of the troops; and although most of, if not
all, our readers are engineers, still it is no disparagement to some of
them to assume that they are more or less unfamiliar with sea water
distillation on the scale on which the process is now being carried on at
Suakim; and as the subject is of general interest, we give a short
description of the process.

In a general sense, fresh water is obtained from sea water by simply
generating steam from the sea water, passing the said steam through a
surface condenser, and filtering the resulting water. The obtaining of
fresh water in this way has been in practice on board sea-going ships for
many years. It is supposed by some authorities on this subject that the
first time fresh water was thus obtained at sea was by an old captain of
a brig which ran short of water, and he cut up some pewter dishes into
strips, which he bent and soldered into a pipe. He, with the carpenter's
aid, fitted a wooden lid in one of the cooking boilers, and fixed one end
of his pipe in it. He next sawed a water cask in half, bored a hole in
the bottom of one half, and took his pipe through it, filling the space
round the pipe with sea water. Thus he extemporized a worm and still or
condenser. The distilled water, however, was scarcely drinkable. Not to
be beaten, however, the captain got some pieces of charred wood which he
put in the water, which so far improved it as to render it at all events
fit to sustain life, and our skipper brought his brig and her screw
safely to port. What suggested the use of charcoal to his mind history
does not tell. For many years past scarce any sea-going vessel leaves
port that is not fitted with a properly constructed distiller; and one
conspicuous advantage attending this practice is that each ship thus
fitted to the satisfaction of the Board of Trade inspector is allowed to
sail with only half the quantity of fresh water on board which she should
have if not provided with a distiller. The distiller and filter occupy
very much less space than that which would be occupied by the casks or
tanks of water otherwise required to be carried.

Coming now a little to detail, sea water distillers are usually fitted in
connection with the winch and its boiler, which latter supplies the steam
both for distillation and to drive the engine working its circulating
pump. Smaller distillers are worked without a pump, the cooling water
merely passing through by gravitation. These smaller affairs again are of
two kinds, the one being mounted at one end of the cooking hearth, as in
outline sketch, which shows a two oven hearth with distiller at one end.
A is the supply pipe to admit air to aerate the water; B is the cock
where fresh water is drawn off; C is a pipe conveying cooling water to
the condenser E, placed on three little feet on top of the boiler, F,
whose steam rises up a central pipe to the dome top, where it expands out
and returns downward through a number of tubes about 1 in. diameter, in
which it is condensed, collected in a bottom chamber, and drawn off
through the cock, B. A distiller of this size would make about thirty
gallons of fresh water per day. Very frequently a distiller, such as is
shown in the sketch, is mounted separately, and placed near the winch or
donkey boiler, which supplies it with steam, the lower part, F, being
then used as a filter. The diameter of E is from 15 in. to 18 in., the
outer casing being either iron or copper. Another form of distiller is
one like the above, but larger, and having a small donkey engine and
circulating pump attached thereto. As a rule these distillers are
vertical, but larger apparatus are arranged horizontally. To give our
readers some general idea of size, weight, and produce of water, we may
say that a plain cylindrical distiller, mounted on a square filter case,
measuring 3 ft. 9 in. high, weighing 41/2 cwt., will distill twelve gallons
per hour. A larger size, measuring 6 ft. 2 in. high, and weighing about
23 cwt., will give 85 gallons; while a still larger one, measuring 7 ft.
high and weighing 32 cwt., yields 150 gallons. These have no pumps. When
an engine and pump are fitted, the weight is increased from about 80 per
cent. in the smaller to 50 per cent. in the larger sizes. An immense
advantage attends the use of those distillers that are combined with a
winch boiler. Of course, the chief use of the winch is while in dock;
some use is made of it at sea to do heavy pulling and hauling, to wash
decks, and in case of emergency the circulating pump is used as a fire
engine. Were it not, however, for the distiller, the winch boiler would
simply be idle lumber at sea. The distiller, however, finds useful
employment for it, and has also this excellent effect, that as steam is
pretty constantly kept up for the distiller, in the evil event of a fire
the boiler is ready to work at once. In horizontal types of distiller an
engine and pump are mounted on a cast iron casing as a bed, and in this
casing is placed a number of tubes through which the steam passes to be
condensed, the whole being simply a surface condenser with engine and
pump above. Another type is that of a small single-flued horizontal
boiler with combustion chamber and twenty or thirty return tubes--in
fact, the present high-pressure marine boiler on a small scale. A boiler
of this sort, measuring 4 ft. to 5 ft. long, 3 ft. 9 in. to 4 ft. 6 in.
diameter, would have a horizontal donkey engine on a bed at its side, and
at the end of the engine a vertical cylindrical condenser.


Few have done more, perhaps none so much, as Dr. Normandy to make sea
water distillation not only a success as a source of water supply, but
also to supply it at a minimum cost for fuel. He by a peculiar
arrangement of pipes embodied something of the regenerative system in his
apparatus, using the heat taken from one lot of steam to generate more,
and again the heat from this he used over again. The defect of his older
arrangements was undue complexity and consequent trouble to keep in

As can be well imagined, the distillers in use at Suakim are on a much
more colossal scale, and owing to the now almost universal use of surface
condensers in ocean steamers, no great difficulty ought to attend the
adaptation of the boilers and condensers of one of our transports. One of
these full-powered steamers will indicate, say, 5,000 horse-power, and
assuming her engines to use 25 lb. of steam per indicated horse-power, or
21/4 gallons, she could distill some 12,000 gallons of water per hour. As
no appreciable pressure of steam need be maintained, the boilers would
suffer little from deposit, especially if regularly blown out. Hard
firing need not be resorted to; indeed, it would be injudicious, as, of
course, priming must be carefully guarded against. Of course, the salt
water distilled will affect the working, not exactly of the distillers,
but of the boilers. If the water in the harbor, as is not improbable, is
muddy, some method of filtering it before pumping it into the boilers
ought, if at all practicable, to be resorted to, for the twofold reason
of preserving the boiler plates from muddy deposit, and also to prevent
priming, which would certainly ensue from the use of muddy water. No
doubt the medical staff take care that the distilled water is alike
thoroughly aerated and efficiently filtered. The most successful method
of aerating is, we believe, to cause the current of steam as it enters
the condenser to suck in air by induced current along with it. The
filtering ought not to present any difficulty, as at all events sand
enough can be had. Charcoal, however, is another affair, and all
distilled water ought to be brought into contact with this substance.

Simple, however, as such an arrangement as this appears to be, practical
difficulties, which it is _said_ are insurmountable, stand in the way of
its adoption, and the distilled water produced for Egypt is made in
special apparatus, and various forms of condenser are employed, made
under various patents. The principle involved is, however, in all cases
the same. Steam is generated in one of the ships' boilers, and condensed,
filtered, and aerated in a special apparatus. The great objection to the
use of the ordinary surface condenser is that the main engines would, in
the majority of cases, have to be kept going, in order to pump the
distilled water out of the condenser, and to supply circulating water.
But it is easy to see that if engineers thought proper, this difficulty
could be readily got over. Separate circulating pumps, usually
centrifugal, are now freely used, and the addition of a special pump for
lifting the condensed water presents no difficulty whatever. While the
main engines are running, the withdrawal of much condensed water would
no doubt risk the safety of the boiler; but in the case of so-called
"distilling" ships, there need be no trouble incurred on this
score.--_The Engineer_.

* * * * *


[Footnote: We take from the Br. Jour. of Photo. the following interesting
paper read by W. Goodwin before the Glasgow and West of Scotland Amateur

With good plates, and intelligent development, a practiced photographer
may within certain limits correct the effects of an over or under
exposure; but you have all, doubtless, found out that there is a correct
exposure, and that you cannot trespass very far on either side of it
without sacrificing something in the resulting negative.

| | Badly lighted| Portraits in bright
| | interiors,| diffused light
Aperture | +------------+ up | out of doors.
calculated | Landscape with | Fairly | to | /
on the | heavy foliage in | lighted | | / Portraits in
standard | foreground. | interiors | | / studio light
system +------+-------+ +------+ | | | /
of the | Sea |Open | | Under| | | | / Portraits
Photographic| and |land- | |trees,| | | | | in ordinary
Society. | sky. | scape.| |up to | | | | | room.
| sec | sec | sec | m s | m s| h m | sec | m s| m s
No. 1, | 1/160| 1/50 | 1/8 | 0 10 | 0 10| 0 2 | 1/6 | 0 1| 0 4
or f/4 | | | | | | | | |
No. 2, | 1/80 | 1/25 | 1/4 | 0 20 | 0 20| 0 4 | 1/3 | 0 2| 0 8
or f/5.657 | | | | | | | | |
No. 4, | 1/40 | 1/12 | 1/2 | 0 40 | 0 40| 0 8 | 2/3 | 0 4| 0 16
or f/8 | | | | | | | | |
No. 8, | 1/20 | 1/6 | 1 | 1 20 | 1 20| 0 16 | 1-1/3| 0 8| 0 32
or f/11.314 | | | | | | | | |
No. 16, | 1/10 | 1/3 | 2 | 2 40 | 2 40| 0 32 | 2-2/3| 0 16| 1 4
or f/16 | | | | | | | | |
No. 32, | 1/5 | 2/3 | 4 | 5 20 | 5 20| 1 4 | 5-1/3| 0 32| 2 8
or f/22.627 | | | | | | | | |
No. 64, | 2/5 | 1-1/3 | 8 |10 40 |10 40| 2 8 |10-1/2| 1 4| 4 15
or f/32 | | | | | | | | |
No. 128, | 4/5 | 2-2/3 | 16 |21 0 |21 0| 4 15 | 21 | 2 8| 8 30
or f/45.255 | | | | | | | | |
No. 256, |1-1/2 | 5-1/2 | 32 |42 0 |42 0| 8 30 | 42 | 4 15|17 0
or f/64 | | | | | | | | |

The estimation of this correct exposure is probably the greatest
difficulty in photography, and it is particularly discouraging to find
plate after plate useless because the guess has been wide of the mark.
There are some here to-night who have spoiled so many plates that at last
they are prepared by experience for almost any contingency, and to those
I nave very little to say; but there are also many who are still in their
troubles, and I propose to tell them how the amount of guesswork required
may be reduced to a minimum.

The factors which govern exposure are: the subject of the picture, the
lens and its aperture, the rapidity of the plate, and last, but not by
any means least, the quality of the light by which the work is to be

Let us consider each of these separately, and see if we cannot reduce any
of them to rule. In this respect the subject will be found somewhat
intractable. Scarcely two subjects will be found to send exactly the same
amount of light through the lens. However, a broad classification may be
made, and this has been done by Mr. Burton in his Table of Comparative
Exposures. A glance at this table will show how greatly the character of
the view may influence the time of exposure. Thus, with full aperture of
a rapid symmetrical, the exposure for open landscape is given as
one-twelfth of a second; when heavy foliage appears in the foreground,
half a second will be required; while, under trees, as much as forty
seconds may be needed.

The first aid I have to suggest is the use of such a table as Mr.
Burton's. Before we do anything more in this direction, we must consider
the influence of the lens and its diaphragms. In theory the single
landscape lens is more rapid than the doublet of equal aperture, but the
difference is so little that it may be disregarded in practice, and my
remarks will apply to both.

The rapidity of a lens depends mainly on its aperture and its focal
length. Thus a lens of twelve inches focus will require four times the
exposure of a six inch, with an equal sized diaphragm, and a quarter inch
diaphragm will require four times the exposure of a half inch when used
in the same lens.

The Photographic Society of Great Britain have recommended that the
diaphragms of all lenses should bear such relation to the focal length
that each should require exactly double the exposure of the next smaller.
Now, if we turn again to Mr. Burton's table, we shall find that it is
constructed on this principle, and that each stop is numbered so as to
show its exposure. Obviously, the most sensible thing would be to get a
set of stops made to correspond with this arrangement, but we will see
how we can construct a table for stops of any size.

First, if possible, find the equivalent focus of your lens. If it is made
by a known maker, you will find it in his price list, and if not, you may
calculate it for yourself by the rules given in the various text books,
provided you have a camera of pretty long focus. However, it will be near
enough for our purpose if you get a sharp image of the sun on a piece of
paper, and while you hold lens and paper, get some one to measure the
distance from the paper to the diaphragm aperture, or, in the case of a
single lens, to the center of the lens. Note down this focal length, and
proceed to measure your diaphragms in sixteenths of an inch.

Then, with pen and paper, proceed to divide the diameter of each stop
into the focus, and state the result as a fraction of the focus, thus
f/8. For example, a Ross half plate rapid symmetrical has a focal length
of 71/2 in.; for convenience reduce this to sixteenths=120. A diaphragm
measuring seven sixteenths will give the fraction f/17. Now let us see if
any of these stops correspond with Mr. Burton's. The first two in his
table will only be found in portrait lenses, but we shall probably find
one to correspond with the third, if we are using a doublet lens; with a
single lens we won't find any so large. Having picked out those that
correspond, and filled in the exposure for them, we have now to deal with
the odd sizes. Here is one, f/27, which is just half way between No. 16
and No. 32, but a moment's thought will show that as the exposure
increases as the square of the diameter, it won't do to take the exposure
half way between the two.

We have another factor to consider now: that is, the rapidity of the
plate. If you use plates by a maker who has a name to sustain, you may be
pretty confident that they are of fairly uniform rapidity, so after you
have got into the way of working any particular brand, the best thing you
can do is to stick to it. The exposures in our table are for plates of
medium rapidity in good spring light. In my own experience I find that
they just suit "thirty times" plates, or fifteen on the sensitometer; but
then I like a full exposure with slow development, and I know that others
find these exposures just right for "twenty times" plates developed in
the usual way. The most rapid plates in the market will not be overdone
with half the given exposures. It must always be borne in mind that an
error of a fraction of a second in either direction may be corrected in
development, and it is impossible to make a very serious error if you
refer to the table.

We come now to the light. If you depend on the eye entirely in judging
the quality of the light, it will sometimes play you tricks. The rays
which are most active on the plates are those which have the least effect
on the eye. We can, however, by chemical means arrive at an exact
estimate of the active power, and for this purpose an actinometer is
used. This is simply an arrangement whereby a piece of sensitized paper
is exposed and allowed to darken to a standard tint, and by the time it
takes to reach that tint the value of the light is judged. Capt. Abney
has, however, pointed out that ordinary sensitized paper is not suitable
for bromide plates, since there are conditions of light in which the
plates will be fairly rapid while the paper will be very slow. He gives a
formula for a bromide paper, which is treated with tannin in order to
absorb the bromine set free during exposure, otherwise the darkening
would be very slight. I used this paper for a while, but found it rather
slow. The tannin also turned brown on keeping for a week or so. I then
made some more, substituting for tannin potassium _nitrite_ (not
nitrate), which is colorless. This was an improvement, but still it was
just slow enough.

However, noticing in Capt. Abney's article the statement that the bromide
of silver should be as nearly as possible in the same state in the paper
as in the plate, I thought "Why not Morgan's paper?" This, of course, is
just bromide emulsion on paper, and if, as I suspect from its color, it
contains a trace of iodide, why, so do most commercial plates. A sheet of
this paper cut into strips, soaked for ten minutes in a fifteen-grain
solution of potassium nitrite, and dried, gives a sensitive paper which
darkens with great rapidity to a good deep tint, and keeps indefinitely.
Here is some prepared last summer, which is still quite good. To use this
paper make a little box so that a little roll of it can be stored in one
end, and drawn forward as required beneath a piece of glass.

Bearing in mind that your table of exposures is calculated for the best
spring light, go to the country some bright day next month with
note-book, actinometer, and the necessary appliances for exposing a few
plates. Select, say, an open landscape, and use your smallest stop. When
all ready to expose, get out your actinometer and expose it to the
reflected light of the sky for ten seconds (if the sun is shining, turn
your back to it, and keep the actinometer in your own shadow); then put
it in your pocket, expose a plate according to your table, and in case
the light or plate should not be just in accordance with the conditions
under which the table was prepared, expose other two plates, one a little
less and one a little more than that first exposed. Then note down
everything you have done--kind of view, stop, speed of plate, exposure of
each plate, and length of exposure of actinometer.

When you get home, the first thing to do is to get hold of a paint box
and paint the underside of the glass of your actinometer to match the
darkened paper. Do this by gas light. Then scrape away a little of the
paint, so as to let a strip of the paper be seen below it. After this
develop your three plates with a developer of normal strength, and see
which is best. If you have chosen a really bright spring day, and are
using plates of medium rapidity, you will most likely find that exposed
according to the table just about right.

Now let us see how we can use these aids in our field work. We have
ascertained the correct exposure with a given stop on one class of view,
with light of a given quality, but now suppose all these conditions
altered. Let the view have heavy foliage coming close up to the camera,
the stop be a size larger than that used in our first experiment, and the
day rather dull. The table tells us what the exposure would be with this
stop on this view, on a bright day; and if the actinometer take twenty
seconds to reach the painted tint, then we must double the exposure given
in the table.

You may sometimes find that the actinometer indicates a very different
exposure from what the eye would lead you to expect. For instance, one
day last September I went to Bothwell Castle, to get a picture I knew of
in the grounds. It was one of those strange yellow days we had then, and
the sun, though shining with all his might, was apparently shining
through orange glass. The actinometer indicated an exposure of thirty
seconds where in good light one would be right. I was rather incredulous.
Thirty seconds in broad sunshine! However, I gave this exposure, but for
my own satisfaction I gave another plate fifteen seconds only.

On developing, the latter was hopelessly underexposed while that having
thirty seconds gave a negative which furnished one of my exhibition

I have shown you how to reduce the quality of the light to a certainty,
also how to reduce to rule the exposure with different lenses and stops
on certain classes of subjects, and it remains with you only to guess
correctly to what class the view you wish to take belongs; I can assure
you from my own experience that there is enough uncertainty about that
point to prevent good negatives ever becoming monotonous.

The only aid I can suggest in this case is the continual use of a
note-book. Note every plate you expose, and when you have a failure be
careful to record the fact, and you will gradually find these accumulated
notes becoming a great help in cases of doubt. One hint I can give to
beginners is that a great number of the pictures to be met with in this
part of the country are intermediate between "Open Landscape" and
"Landscape with heavy foliage in foreground;" and it is scarcely needful
to say that if you are in doubt, let the exposure be rather too much than
too little; you _may_ make a negative of an overexposed plate, but never
of an underexposed one.

* * * * *


[Footnote: Read at the stated meeting of the Franklin Institute, March
18, 1885.]


It is well known that the ordinary photographic processes do not
reproduce colors in the true proportion of their brightness. Violet and
blue photograph too light; green, yellow, orange and red, too dark. For a
long time it was believed to be impossible to remedy this defect; and
even when it became known that bromide of silver could be made more
sensitive to yellow and red by staining it with certain dyes, the subject
received very little attention, because it was also known that the
increase of sensitiveness was too slight to be of practical value in
commercial photography.

Dr. H.W. Vogel, who was one of the first, though not the first, to devote
attention to this subject, announced, in 1873, that he had succeeded in
making a yellow object photograph lighter than a blue or violet one, by
using a silver-bromide plate stained with coraline, and exposed through a
yellow glass. The plate showed no increased sensitiveness to red, and the
experiment, although of considerable scientific interest, did not
indicate a practically useful process.

In the spring of 1878 I became interested in this subject, and tried to
discover a method of producing plates which should be sensitive to all
colors, and capable of reproducing them in the true proportion of their
brightness. I commenced by trying nearly all the color sensitizers which
had already been suggested, in order to learn which was the best, and
then, if possible, _why_ it was the best, as a guide to further research.
Chlorophyl was the only thing I tried which was sufficiently sensitive to
red to offer any encouragement in that direction; but the solution which
I obtained was weak and unstable, and far from being a satisfactory color
sensitizer. Hoping to obtain a better solution with which to continue my
experiments, I made extracts from many kinds of leaves, and found that a
solution from blue myrtle leaves looked better and kept better than any
other, and when it was applied to the silver-bromide plates they became
remarkably sensitive, not only to all shades of red, but also to orange,
yellow, and green. By placing in front of the lens a color-screen
consisting of a small glass tank containing a weak solution of bichromate
of potash, to cut off part of the blue and violet light, I obtained, with
these chlorophyl plates, the first photographs in which all colors were
reproduced in the true proportions of their brightness. But my chief
desire at that time was to realize a method of producing from any object
in colors a set of three negatives, in one of which the shadows should
represent the blue of the original, in another the yellow, and in another
the red, in such a manner that transparent pigment prints from these
negatives--blue, yellow, and red--would, when superimposed on a white
surface, represent not only the lights and shadows, but also the colors
of the object. This had already been attempted by others, who failed
because their plates were not sufficiently sensitive to red and yellow.

Having succeeded perfectly in my undertakings, I published my discovery
in 1879,[1] explaining how to prepare and use the chlorophyl plates, in
connection with the yellow screen, for the purpose of securing correct
photographs of colored objects.[2]

[Footnote 1: _Philadelphia Photographer_, December, 1879, p. 365.]

[Footnote 2: I intended this publication to be a very full and explicit
one, and it was sufficiently so to be perfectly understood by most who
saw it; but some may think I did not sufficiently emphasize the
importance of using the particular kind of chlorophyl which I mentioned.
In a brief communication to the editor of the _Photo. News_, in 1883, I
described some experiments with eosine as a color sensitizer, and then
called attention to the superiority of blue-myrtle chlorophyl for this
purpose, stating that I had not been able to secure such results with any
other kind of chlorophyl, and that a fresh solution from fresh leaves
must be used to secure the greatest possible degree of sensitiveness. See
_Photo. News_, Nov. 1883, p. 747.]

So far as I know, nobody tried the process. Nearly five years later Dr.
Vogel announced that, after eleven years of investigation, he had at last
realized a successful process of this character, and that this new
process of his was the "solution of a problem that had long been
encompassed with difficulty." This publication attracted a great deal of
attention, and gave me occasion to again call attention to my process,[3]
and point out that it was not only the first practical solution of this
problem, but the only truly isochromatic process ever discovered. Dr.
Vogel's new process was not only no better in any respect, but the plates
were insensitive to scarlet and ruby-red, and therefore would not
photograph all colors in the true proportion of their brightness.

[Footnote 3: _Photo. News_, London, September 5, 1884, p. 566, and _Year
Book of Photography_ for 1885, p. 111.]

My method consists in treating ordinary collodio-bromide emulsion plates
with blue myrtle chlorophyl solution, exposing them through the yellow
screen, and then developing them in the usual manner. The emulsion which
I have employed is made with an excess of nitrate of silver, which is
afterward neutralized by the addition of chloride of cobalt; it is known
as Newton's emulsion. I now prepare the chlorophyl from fresh blue myrtle
leaves, by cutting them up fine, covering with pure alcohol, and heating
moderately hot; the leaves are left in the solution, and some zinc powder
is added, which helps to keep the chlorophyl from spoiling. I have a
bottle of this solution which was prepared about six months ago, and now
appears to be as good as when first made.[4] A glass plate is flowed with
the emulsion, and as soon as it has set, the chlorophyl solution is
applied for a few seconds, after which the plate is washed in pure water
until smooth, when it is ready for exposure.

[Footnote 4: I originally recommended chlorophyl extracted from dried
leaves, because I had not yet learned how to preserve the solution for
more than a few weeks; and at some seasons it would be difficult, if not
impossible, to obtain fresh leaves. The tea organifier which I
recommended is also a color sensitizer, and when it is used in connection
with the chlorophyl from dried leaves the plates are as sensitive to red
as can be safely prepared and developed in the light of an ordinary
photographic "dark-room." Plates prepared with chlorophyl from fresh
leaves do not require treatment with the tea organifier to secure this
degree of sensitiveness. Recently I have used the tea organifier and some
other sensitizers, in connection with the solution from _fresh_
myrtle-leaves, and in this way have produced plates having such an
exalted color sensitiveness as to be unmanageable in ordinary "dark-room"
light. Possibly, such plates might be prepared and developed in total
darkness, by the aid of suitable mechanical contrivances, but I am not
sure that they would work clear even then, because they appear to be
sensitive to heat as well as to light.]

My color-screen consists of a small plate-glass tank, having a space of
3/16 of of an inch between the glass, filled with a solution of
bichromate of potash about one grain strong. I place the tank in front of
the lens, in contact with the lens-mount. The advantage of this tank and
solution is that it can be more easily obtained than yellow plate glass,
and the color can be adjusted to meet any requirement.

The plates require about three times as much exposure through the yellow
screen as without it, and may be developed with the ordinary alkaline


In order to illustrate the value of this process, I made two photographs
of a highly-colored chromo-lithograph, representing a lady with a bright
scarlet hat and purple feather, a yellow-brown cape and a dark-blue
dress. One, by the ordinary process, represents the blue as lighter than
the yellow-brown, the bright scarlet hat as black, and the purple feather
as nearly white. The other, by the chlorophyl process, reproduces all
colors in nearly the true proportion of their brightness, but with a
slight exaggeration of contrast produced purposely by using a too-strong
color solution in the small tank.

I also made two landscape photographs, one by the ordinary process, and
the other by the chlorophyl process, exposing them simultaneously. In the
ordinary photograph, distant hills are lost through overexposure, yet the
foreground seems underexposed, and yellow straw-stacks and bright autumn
leaves appear black. In the chlorophyl photograph, the distant hills are
not overexposed, nor is the foreground underexposed; the yellow
straw-stacks appear nearly white, and bright autumn leaves contrast
strongly with the dark green about them.

To test the relative color-sensitiveness of plain emulsion plates, plates
stained with eosine, and plates stained with the blue-myrtle chlorophyl,
I exposed one of each kind through the same yellow screen, giving each
five minutes exposure, on the same piece of copy, which was the
chromo-lithograph already described. The plain emulsion plate showed only
the high lights of the picture, after prolonged development. The eosine
plate was underexposed, but brought up everything fairly well except the
scarlet hat, which came up like black. The chlorophyl plate was
overexposed, brought out all colors better than the eosine plate, and
gave full value to the bright scarlet of the hat, the detail in which was
beautifully rendered.

Dr. Vogel advanced the theory that silver-bromide is insensitive to
yellow and red, because it reflects or transmits those colors; and that
it becomes sensitive when stained, because of the optical properties of
the dyes. He afterward admitted that only such dyes as are capable of
entering into chemical combination with the silver-bromide proved capable
of increasing its sensitiveness to color, but he held to the theory that
the optical properties of the compound were the cause of its

I have shown that the color-sensitiveness can be produced by treatment
with an organic compound which has none of the optical properties
characteristic of dyes; and that chlorophyl, which absorbs only red
light, greatly increases the sensitiveness also to yellow and green.
There is, therefore, good reason to doubt if the color-sensitiveness is
ever due to the optical properties of the dye or combination.

Attempts have been made to produce isochromatic gelatine dry plates
which, while many times more sensitive to white light than my chlorophyl
plates, shall also show the same relative color-sensitiveness. Such
plates would be very valuable but for one fact: it would be necessary to
prepare and develop them in almost total darkness. Gelatine bromide dry
plates extremely sensitive to yellow, but _comparatively insensitive to
red_, might be used to advantage in portrait and instantaneous
photography, because they could be safely prepared and developed in red
light; but when truly isochromatic photographs are required, the time of
exposure must be regulated to suit the degree of sensitiveness to red,
which cannot safely be made greater than I have realized with my
chlorophyl process.

* * * * *


The effect of the unequal expansion of paper, when wetted, in causing
distortion of the photographic image impressed upon it, has, in the case
of ordinary photographs upon albumenized paper, been well recognized; but
the extent to which such distortion may exist under different treatment
is worthy of some special consideration, particularly with reference to
the method of printing upon gelatinized paper, which has been thought by
some likely to supersede the method now usually employed with albumenized

When a print upon the ordinary photographic (albumen) paper is wetted,
the fiber expands more in one direction than in the other, so that the
print becomes unequally enlarged, very slightly in one and much more so
in the other way of the paper. When the paper is dried without any strain
being put upon it, the fibers regain very nearly their original
dimensions and position, so that the distortion which has existed in the
wet condition nearly disappears.

If the photograph is cemented, while in the expanded condition, upon a
rigid surface, the distortion then existing is fixed, and rendered
permanent. Such a cementation or method of mounting is that which has
been generally adopted, and the consequence has been that every now and
then complaints have justly been made of the untruthfulness--owing to
this particular distortion--of photographs; productions whose chief merit
has often been asserted to consist in their absolute truthfulness. This
distortion is very manifest when, in a set of portraits, some of the
prints happen to have been made in one direction of the paper, and others
with the long grain the other way. I have known a case where a proof
happened to increase the face in width, and all the other prints
increased it in length. Of course, neither was correct, but the proof had
been accepted and liked, and the remainder of the set had to be reprinted
with the grain of the paper running in the same direction as that in the
first one which had been supplied.

Another evil arising from mounting prints while expanded with moisture
is, that in drying the contraction of the paper pulls round the card into
a curved form and although by rolling this curvature may be temporarily
got rid of, the fiber of the paper is in a strained condition, and the
bent state of the mount is, sooner or later, renewed thereby.

To remedy these evils it has been proposed to mount the print when dry,
by forcible pressure against a slightly damped card, the back of the
print having been previously coated with a cement and dried. This plan
is, to a great extent, successful; but that it does not give absolute
immunity from distortion is, I think, evident from the following
consideration. The prints, after being mounted a few days, will show a
certain tendency to curl inward. This curling, I take it, is a measure of
the strain upon the print, produced by the more complete return to its
original dimensions of the paper photograph. Probably it would be well to
keep the prints a few days after drying, or to subject them to
alternations of damp and dryness, in order to facilitate this complete
return before being placed upon the card. The evil of distortion is,
however, very slight--perhaps imperceptible--compared with that existing
when the prints are mounted wet. I may mention, _en passant_, that I have
found gum much more satisfactory as a mountant than starch paste in what
is known as the "dry mounting" system.

The paper which has recently been introduced for producing prints by
development upon a gelatine surface does not generally, when dried in the
usual way, give so good or so brilliant a surface as that of albumenized
paper; but on the other hand it is very easy with it to obtain what is
called an enamel surface, by simply allowing it to dry in contact with a
prepared surface of glass. This method of finishing has therefore been
much recommended and adopted, but without consideration of the effect of
distortion in connection with it. In an ordinary photograph the print is
mounted damp, but in the case of a print squeegeed on to the glass, the
paper is saturated and thoroughly swollen, and the use of the squeegee
strains it out to its fullest extent. By drying in the position in which
it has been held by contact with the glass, the distortion becomes fixed,
and if the print is mounted while in this state the distortion is made
permanent. How long the strain and distortion remain in an unmounted
print, and whether by time and alternations of moisture and dryness the
strain would be lost, and if so, whether the brilliant enamel surface
would go at the same time, are questions worthy of further investigation
and discussion.

For mounting prints upon developed gelatine paper, it has been
recommended to cement the edges only, so as to leave the greater part of
the print with its enamel surface. This plan is unsatisfactory, for two
reasons, besides the objection on the ground of distortion. There is a
rough-looking margin which spoils the continuity of appearance,
especially (as in the specimens I have seen) where the line of cement is
not kept at an exact width, but encroaches here and there.

Secondly, the print, from not being attached to the mount all over, is
apt, especially when in a large size, to be somewhat wavy and wanting in
flatness. Another plan recommended, as giving a surface resembling
albumen paper, is to paste the back of the print without moistening the
surface, and so mount. Some prints that have been shown thus treated had
so strongly curled the cards upon which they were mounted that it is
evident there was considerable strain and consequent distortion.

A third plan recommended is to paste the back of the print while in
contact with the glass upon which it has to dry; and, when dried, to
mount by passing through a rolling press with a damped card. This plan
looks, at first sight, like that recommended for albumen paper, and
called "dry" mounting. Consideration, however, will show that there is a
radical difference. In the case of the albumen paper the print has been
dried without strain, and therefore but little change is to be looked
for, while the print dried in contact with glass is strained to the
utmost, causing present distortion and future curling of the mount.
Perhaps the evil of distortion caused by enameling may be reduced to a
minimum by soaking the print in alcohol previous to laying it upon the

Since the distortion of the photograph arises from the unequal expansion
of the paper when wet, it becomes a question whether something may not be
done in the selection of the paper itself. It may be that some makes vary
much less than others in the "length against width" extension of the
surface by wetting. It must be remembered that for gelatine emulsion we
are not nearly so limited in the selection of paper as when it is
required to be albumenized. In the latter case the image is in the paper,
whereas with gelatine the image is contained in the surface coating. I
may mention that the best plain, i.e., not enameled, but resembling that
of ordinary albumen paper, surface that I have seen upon gelatine paper
was upon some foreign post that I had obtained for another purpose. The
emulsion employed was that described by Mr. J.B.B. Wellington, and this
gentleman agreed with me in attributing the superiority of the surface
obtained to the fine quality of the paper upon which the emulsion had
been coated. Some commercial samples appear to be coated upon paper of
somewhat coarse texture. This does not show when the print is enameled.

The unequal expansion of paper is a subject of interest, not only in
connection with gelatine paper for development, but with various
photographic processes. In making carbon transparencies for instance, the
gelatine film which is squeegeed against the glass necessarily takes its
dimensions from the paper to which it is attached, and if that be
expanded more in the one direction than another, the transparency is
similarly deformed; and so, of course, is any negative, enlarged or
otherwise, produced in the camera therefrom. A reproduced negative by
contact printing may either have the distortion due to expansion of the
paper bearing the gelatine film removed or doubled, according to the
direction in which the paper is used for the new negative.--_W.E.
Debenham, in Br. Jour. of Photography_.

* * * * *


An ingenious process for determining the thickness of iron plates in
boilers, or places where they cannot otherwise be measured without
cutting them, has been invented by M. Lebasteur. He spreads upon the
plate the thickness of which he desires to find, and also upon a piece of
sheet iron of known thickness, a layer of tallow about 0.01 inch thick.
He then applies to each, for the same length of time, a small object,
such as a surgeon's cauterizing instrument, heated as nearly as possible
to a constant temperature. The tallow melts, and as in the thicker plate
the heat of the cautery is conducted away more rapidly, while in the thin
plate the heat is less freely conducted away, and the tallow is
consequently melted over a large area, the diameters of the circles of
bare metal around the heated point, bounded after cooling by a little
ridge of tallow, will be to each other inversely as the thickness of the
plates. The process is stated to have given in the inventor's hands,
results of great accuracy.

* * * * *


The pediment of the central pavilion and the two side pavilions of the
new House of Parliament, at Vienna, are to be ornamented with groups of
statuary. The group in the middle pediment represents the granting of the
constitution by the Emperor Francis Joseph, and was executed by Professor


The pediment of the left wing is ornamented by a group representing
Justice, and the pediment of the right wing by a group representing the
Home Government.

Johannes Benk, the well known Austrian sculptor, designed and executed
the last mentioned group. The two figures at the left hand end of this
group represent Science and Literature, and those at the right hand end,
Industry and Commerce. The entire group consists of nine figures, the
middle figure being seated and the rest standing, sitting, and lying, as
the space in the pediment allows.

A seated female figure studying a papyrus roll represents Science, and
the adjacent female figure, resting one arm on the figure representing
Science, and the other, on a lyre, represents Literature or Poetry.

Industry is represented by a strong and powerful woman holding a hammer,
and the figure of Mercury and the prow of a vessel represent Commerce.


The modulation and formation of each figure conform strictly to Grecian
models, as does also the entire arrangement of the figures in the group;
and yet there is much of modern life in the figures, especially in the
faces, in which the stereotyped Grecian profile has not been adopted. The
attitudes of the figures are also freer and more easy than those of the
Grecian period.--_Illustrirte Zeitung_.

* * * * *


[Footnote: Paper read before the American Association for the Advancement
of Science, at Philadelphia, Sept, 5, 1884.]

By C.E. FRITTS, 42 Nassau St., New York, N.Y.

In all previous cells, so far as I am aware, the two portions or parts of
the selenium at which the current enters and leaves it have been in
substantially the same electrical state or condition. Furthermore, the
paths of the current and of the light have been transverse to each other,
so that the two forces partially neutralize each other in their action
upon the selenium. Lastly, the current flows through not only the surface
layer, which is acted upon by the light, but also the portion which is
underneath, and not affected thereby, and which therefore detracts from
the actual effect of the light upon the selenium at the surface.

My form of cell is a radical departure from all previous methods of
employing selenium, in all of these respects. In the first place, I form
the selenium in very thin plates, and polarize them, so that the
opposite faces have different electrical states or properties. This I do
by melting it upon a plate of metal with which it will form a chemical
combination, sufficient, at least, to cause the selenium to adhere and
make a good electrical connection with it. The other surface of the
selenium is not so united or combined, but is left in a free state, and a
conductor is subsequently applied over it by simple contact or pressure.

During the process of melting and crystallizing, the selenium is
compressed between the metal plate upon which it is melted and another
plate of steel or other substance with which it will not combine. Thus by
the simultaneous application and action of heat, pressure, chemical
affinity, and crystallization, it is formed into a sheet of granular
selenium, uniformly polarized throughout, and having its two surfaces in
opposite phases as regards its molecular arrangement. The non-adherent
plate being removed after the cell has become cool, I then cover that
surface with a _transparent conductor of electricity_, which may be a
thin film of gold leaf. Platinum, silver, or other suitable material may
also be employed. The whole surface of the selenium is therefore covered
with a good electrical conductor, yet is practically bare to the light,
which passes through the conductor to the selenium underneath.[5] My
standard size of cell has about two by two and a half inches of surface,
with a thickness of 1/1000 to 5/1000 inch of selenium. But the cells can,
of course, be made of any size or form. A great advantage of this
arrangement consists in the fact that it enables me to apply the current
and the light to the selenium in the same plane or general direction,
instead of transversely to each other as heretofore done, so that I can
cause the two influences to either coincide in direction and action, or
to act upon opposite faces of the selenium and oppose each other,
according to the effect desired.

[Footnote 5: The method of constructing the cells was described in the
SCIENTIFIC AMERICAN SUPPLEMENT, No. 462, for Nov. 8, 1884, page 7371.]

By virtue of the process and arrangement described, my cells have a
number of remarkable properties, among which are the following:

1. _Their sensitiveness to light_ is much greater than ever before known.
The most sensitive cell ever produced, previous to my investigations, was
one made by Dr. Werner Siemens, which was 14.8 times as conductive in
sunlight as in dark. In table A, I give results obtained from a number of
my cells.

It will be observed that I have produced one cell which was 337.5 times
as conductive in hazy sunlight as in dark. The tremendous change of
resistance involved in the expression "337.5 times" may perhaps be more
fully realized by saying that 99.704 _per cent_. of the resistance had
disappeared temporarily, under the joint action of light and electricity,
so that there remained _less than 3/10 of 1 per cent_. of the original
resistance of the selenium in dark.

In order to obtain these high results, the cells must be protected from
light when not in use. The resistance is first measured while the cell is
still in total darkness. It is then exposed to sunlight and again
measured. It is also necessary to send the current in at the gold
electrode or face, as the cell is much less sensitive to light when the
light acts upon one surface of the selenium and the current enters at the
opposite surface. When the two influences, the light and the current, act
through the gold, in conjunction, their forces are united; and, as every
atom of the selenium is affected by the light, owing to the extreme
thinness of the plate, we have the full effect shown in the measurements.



Selenium | Battery | Resistance in | Resistance in |
cell. | power. | dark. | sunlight. | Ratio.
| | ohms. | ohms. |
No. 22 |5 elements.| 39,000 | 340 |114 to 1
" 23[6]|5 " | 14,000 | 170 | 82.3 " "
" 24[7]|5 " | 648,000 | 2,400 |270 " "
" 25 |5 " | 180,000 | 930 |196.5 " "
" 26 |5 " | 135,000 | 710 |190 " "
" 107 |5 " | 118,000 | 740 |159 " "

Book of the day:
Facebook Google Reddit StumbleUpon Twitter Pinterest