object from a slightly different angle. Mr. Ives devised a beautiful solution of that elegant simplicity characteristic of his work in general.

In Fig. 6 the camera body A is of the box type, with the usual lens system B. The plate-holder is carried at the back in the usual way, but carries two plates, C and D, the rear one having its film side nearest the lens, and the one in front of this having its film face to face with the rear one. A third plate, E, lies in the bottom of the camera. Light from the object passes through the lens system, and a compensating color screen, F, immediately behind the lens system. The light next strikes a clear glass, G, that is placed at an angle of forty-five degrees. The front face of this glass acts as a mirror to deflect

part of the light to the bottom plate, E. The remaining light passes through the glass G, through plate D, to act on the sensitive rear face of D and on the sensitive front face of plate C. It is clear that a single exposure suffices for making the three plates, C, D, and E. Each of these plates must, of course, select its particular primary color. The principle under which they do this is old, consisting of the filtering out by suitable interposed color filters. These color filters are located as follows: The compensating screen F acts also as a color filter; a second one is supplied by a varnish on the back of the inclined transparent mirror, G, and, finally, the sensitive film of D is suitably dyed to filter for plate C. Much work and patient

investigation and experimenting were necessary to secure the proper relationship of the sensitive plate emulsions, different for each of the three plates and of the color filters. Some of the difficulties may be imagined when it is realized that, even though all the light filters were of proper color and purposely balanced as to intensity of color, an emulsion slightly too fast or too slow on one plate would result in that plate being too dense or too thin, and so give not enough or too much of its color in the final result.

The arrangement of the three plates for convenient handling is also worthy of note. All three plates are attached at their lower edges, H, to a strip of gummed paper, on which they hinge like the leaves of a book. They are inserted as a unit into the plate-holder, which is slid into the camera back in the usual way. At this time the mirror, G, is held up out of the way at the top of the camera by a simple catch. When the plate-holder slide is withdrawn, that allows the first plate, E, to fall downward to its position, as shown in Fig. 6. Plates C and D cannot fall, being slightly larger than the front opening of the holder. The mirror, G, is then dropped to place, and the exposure made precisely as with any ordinary camera. This camera can, in fact, be used for ordinary work by simply leaving the mirror, G, at the top, and using ordinary single plates, roll films, or film packs. Ordinary cameras of the box type may also be readily converted for color work by adding the mirror G and slightly adapting the plate-holder slide.

The plates are developed in the usual way; the best method is by fixed time development in a standard solution, glycin preferred. A convenient arrangement is also due to Mr. Ives, consisting of a simple tank adapted to take the triple plate pack ("tripak"), with the three plates spread apart like the leaves of a book, for convenient access by the developer. Only a single developer is required, acting five to eight minutes, and the ordinary fixing in hypo and water washing. To print out, all three plates are placed side by side in a long printing-frame and the film exposed. The film is gelatin bromid, carried on a very thin celluloid backing. When the proper exposure has been made, by the aid of a very simple tint exposure meter, the film is washed out in hot water until it is clear, and then passed through a hypo. The film is now cut apart, and the three films laid each into its dye-bath of peacock-blue, magenta, and yellow respectively. In about five minutes each will have absorbed the proper amount of dye. As the absorption

power of the film and the dye-baths are all standardized for proper relationship, a more prolonged dyeing has no appreciable effect, and neither skill nor judgment as to proper dye density is needed. The dyed prints are dried in the usual way, placed in register over one another, and bound securely at the edges by passe-partout paper. For lantern slides they are also bound between clear glass plates. Obviously, any number of transparencies may be made from one set of plates. The entire process is exceedingly simple, and, owing to the thorough standardization of all the elements for themselves and in their interrelation, the results are uniform and independent of special manipulative skill or judgment.

This whole chapter of progress on the production of correct color transparencies in any quantity from a single exposure may now be said to be beyond the realm of the laboratory, and to have definitely arrived at that stage of perfection required for broad general public use. With it Mr. Ives has added to his achievements as the originator and perfector of the half-tone printing process and of the three-color printing process the further one of direct practical three-color transparency photography.

With the characteristic energy of the truly scientific inventor, Mr. Ives is already hard at work on the conquest of new worlds, and I have been privileged to see very nearly perfected color photographic prints for direct vision. I trust that it may be my privilege to have the final result brought before the world for the first time through this Club.*

A number of slides were shown illustrating various color phenomena, as color fringes, interferences, bands, etc.

PAPER NO. 1107.

PROPULSIVE MACHINERY AND OIL FUEL IN THE
UNITED STATES NAVAL SERVICE.

CAPTAIN C. W. DYSON, U.S.N.

(Visitor.)

Read January 6, 1912.

UNTIL within the last few years the improvements in propelling machinery for naval vessels, and for marine purposes in general, were few, the designers apparently considering that the reciprocating engines which they were then using were good enough, and that any further improvements in them could be made only at an undesirable increase in weight, in cost, and in complication.

This apparent view extended not only to the main propelling machinery, but also to the auxiliary machinery, with the result that each new ship was practically a copy of those that preceded it, only such modifications being made as the necessities of the particular case dictated.

With the advent of the turbine it became necessary, in case the reciprocating engine was to hold its own, to take advantage of every possible opening for improvement. Such improvements as appeared possible at the time of laying down the designs were made, and that they were desirable has been amply shown by the results on trial and in service obtained by the U.S.S. South Carolina, Michigan, and Delaware.

In the adoption of turbine machinery the Navy Department proceeded with characteristic caution, and no designs of this type of machinery were laid down until results obtained abroad were of such nature as practically to insure success.

In the fall of 1904 it was determined to lay down three scout cruisers, Birmingham, Salem, and Chester, and in order to obtain. data for use in future designs it was decided to fit the Birmingham with reciprocating engines, the Salem with Curtis turbines, and the Chester with Parsons turbines.

Before these vessels were completed, in June, 1907, the Fore River Shipbuilding Company completed and tried out the Southern

Pacific Steamer Creole, which was fitted with Curtis turbines as main propelling engines. The trials were witnessed by representatives of the Navy Department, who, entirely discounting the fact that such a vessel as the Creole was unfitted by her hull and slow speed for turbine propulsion, reported on the turbines only, and after pointing out defects which could be corrected, stated that the Curtis turbine was adapted for marine propulsion.

During this same year, 1907, the designs for the battleships Delaware and North Dakota were being prepared. When the advertisement for bids for these two vessels were issued, bids were invited both for reciprocating engines and for turbines. The design of these vessels made it impossible to install satisfactory Parsons turbines, but single-unit Curtis turbines fitted in very well. Upon the opening of the bids the Bureau of Steam Engineering pointed out to the Navy Department that if turbine machinery were installed in either of these vessels, it would be at the sacrifice of cruising radius. How well based this criticism was has since been amply demonstrated by the performances of the Delaware with reciprocating engines, the North Dakota with Curtis turbines, and the Utah with Parsons turbines.

PROPELLING MACHINERY OF FAST AND LIGHT VESSELS.

For main propelling engines for such vessels as destroyers and torpedo-boats the steam turbine was welcomed with open arms. For such vessels, having high speed, high power, and very light foundations for the support of the propelling machinery, a very high piston speed was necessary, with the reciprocating engines which had formerly been fitted. This high piston speed, with the consequently high number of reversals in the direction of motion of the moving parts, resulted in the production of excessive vibrations of hull and machinery, and the resultant breakdowns of the engines at full power were numerous.

By the adoption of turbines, with the consequent change from reciprocating to rotary motion, the problem of engine vibration was immediately solved, and the only vibration now existing in such vessels is that due to the propellers.

The first turbine-propelled destroyers, Nos. 17 to 21, five in number, were laid down in 1906, and all the destroyers designed, built and building, since that year-a total of 29-have been fitted with turbines. Some of these vessels have the Parsons turbine, with the power distributed between three shafts, while others have the Curtis or the Zoelly, with the power distributed on two shafts.