THE FOLLOWING NOTES relate to the two part article on making objective lenses by the test-plate method, by Patrick A. Driscoll of Rochester, in the March and April numbers.

In "A.T.M." page 71, the middle paragraph suggests a principle known as the "unit focus" principle: all the pertinent design data-focal length, diameters, thicknesses, radii of curves- in this instance of an eyepiece, may be divided or multiplied by any amount chosen (of course with due consideration to practical limitations), provided this remains throughout the same amount. These changes do not, of course, change the focal ratio.

When this fact was called to Driscoll's attention as a possibility for use in connection with his lens specifications he replied: "Now that you dig up one of the deep 'secrets' of our optical wizards that the amateur is not supposed to be let in on, I may as well let the cat out of the bag. An achromatic doublet can very easily be reduced in this way. For example, a 6" can be to reduced to a 3", simply by dividing everything-except, of course, index and dispersion-by 2. It is as easy as that. I didn't put it in my story because if I'm makin' a lens, then by jeepers I'm makin' a big one-which by this method is as easy as a little one-and then the finished article is a man-sized hunk of apparatus. But I suppose the monetary outlay for large blanks is a consideration. If you reduce my 5-1/4" to, say, 3" or so, you reduce the aberrations; also the range of the figure of each surface is not quite so critical and the performance will be slightly better; but-a warning-do not increase the diameter unless willing to accept an increase of these same characteristics. At least, if you risk this, the surfaces must be very exact, else performance will suffer."

In weighing the matter of size, "A.T.M.A.", page 192, bottom paragraph, should not be overlooked.

This unit focus opportunity confers an added degree of freedom on the test-plating procedure, and if now we amateurs can between us collect and pool a variety of specifications for tried and good objective lenses, this method will enjoy still more freedom. Incidentally, there also is nothing about this method which forbids anyone from designing his own objective, just as in the other method, and in the most rigorous way if desired, yet it does make available to those who aren't too happy when tussling with design (as a few aren't, according to your scribe's mail) a chance to make and enjoy their own refractors. By this method the amateur's fear of the convex surface also is resolved, and it enjoys other advantages.

Specifications for two types of cemented and two of uncemented objectives, given on the unit focus basis, appear on page 29 of the newly published book, "Telescopes and Accessories," by Dimitroff and Baker. (This is the book which was given an advance review on page 466 of the 1941 printing of "A.T.M." - it has finally been published.)

Driscoll's specifications call for objectives of rather short focus (large focal ratio); that is f/12, and f/13. Ellison calls for f/15 or f/16 and so do some others. Questioned about this, Driscoll answered: "The 5-1/4" was computed for me as an ideal astronomical glass that would utilize easy c and f and not require too precise adherence to stated tolerances." Dr. D. Everett Taylor has made one and has reported it good, in the April number.

Driscoll's method of calipering the edge of the lens elements during preliminary work-tipped micrometer anvils and stop block-happily complements Ellison's always questionable if not inadequate method in "A.T.M." (pages 115, 117), for it is otherwise almost impossible to caliper a tapered surface twice alike to close tolerance. In "A.T.M." (page 250) Haviland describes an edge gage involving three steel balls and a dial gage (a similar rig is used at the Bureau of Standards and reads to 0.0002" as the lens is rotated in it, according to Ferson), and in Scientific American, February, 1940, D. Everett Taylor also described a very beautiful combined edge thickness gage and spherometer. Both of these contribute in large measure to precise work.

Any who give Driscoll's procedure or program a workout are urged to report all their experiences, impressions, opinions, and findings-its best, its worst. It is hoped that this flatless, test-plating method will result in a larger proportion of refractors being made.

THE REFLECTOR is a splendid telescope but in recent years more and more amateurs have quietly expressed their conviction, gradually arrived at, that on balance the refractor is even more satisfactory, giving steadier, better resolved, sharper, even if less brightly illuminated, images that stand higher magnification, especially for planetary and lunar observation. Here are some recent comments from amateurs who seem happy about their new refractors.

"I still use my 10" Cassegrainian, but the second-hand 5-1/2" Brashear refractor I just bought has so much better definition and resolving power that there is little point in using the Cassegrainian. Almost any night I can use the quarter-inch eyepiece."

Another: "I have just finished a 2-3/4" objective lens, my first, and from here on I am on refractors. This telescope defines perfectly."

A third: "Refractors at times throw theoretical resolving power out the window."

Of course, these are warm statements reflecting new enthusiasms, and they may be subject to some discount. The reflector is an excellent telescope but the notion that there is something inferior about the refractor is all wrong. Maybe 'twas a rationalization because making one called for a lath and a flat. Pat Driscoll's program now, lops off the flat.

CERIUM OXIDE is now strongly competing with rouge in the optical industry. The following notes on that new polishing abrasive are by Frank Allen Lucy, 3427 W. Penn St., Philadelphia, Pa., who has contributed to this department before (July, 1943-notes and equations on Gee's method of figuring Cassegrainian secondaries by the test-plate method):

"People in increasing number have been recommending ceria (ceric oxide) in place of rouge for optical polishing. On the credit side, it is said to be cleaner and faster than rouge, less likely to cause sleeks, turned edges, or tarnishing on aging. On the debit side, it is said to cut so fast that it makes delicate figuring difficult, and to give a dull surface at first, which brightens only on long polishing; so that some think it unsuitable for final figuring, in which the periods are necessarily short.

"The writer's observations on the use of ceria may interest other optical workers. Ceria, being practically white, is certainly cleaner than rouge. Minimizing turned edge seems a matter more of technique than of agent. The commercial sample in our laboratory contains particles ranging widely in size. Probably, if used as received, it would give a dull surface at first, until the coarser particles were well driven into the pitch of the lap, after which the surface would brighten. However, when shaken with water and a wetting agent (Duponol D happened to be the readiest to hand), this ceria was readily dispersed and levigated. The levigated material gives a brilliant polish on any glass from dense barium crown to Pyrex, and works, in my estimation, about twice as fast as rouge of similar scratchiness.

"The scratches here referred to are those necessary ones which remove material in polishing, the scratches which fuzz the edges of a knife-edge shadow. (A. W. Everest, on page 22, 'A.T.M A.,' says, 'Also decide right here what you are going to call the shadow; probably the point where none of the remaining spiderwebs of light crosses the pin.' These spiderwebs are diffraction patterns of scratches, scratches usually too small to be seen by direct examination but nevertheless capable of diffracting light.)

"Polishing speed was judged by the length of spells necessary to produce a given rate of approach to an aspherical figure, starting from a sphere. Thus, if 40 minutes with rouge produced 5 percent of the total correction, 20 minutes with ceria was found adequate, under the same conditions, to produce another 5 percent. Other workers have stated that ceria is faster than rouge by factors ranging from 1.3 to 4 times. Particle size and possibly shape, hardness of pitch, polishing speed, and pressure, proportion of water, rate of feed, and so on, all affect the rate at which glass is removed, making a truly scientific evaluation of this factor difficult. Further, a set of conditions which gives the best results with a given rouge sample is not the best set for any other rouge sample, and is most unlikely to be best for a given ceria sample. The consensus is, however, that properly handled ceria is decidedly faster than properly handled rouge.

"Some hold that ceria is good for coarse work but not for fine. To this, the writer cannot subscribe. For example, Dévé states, 'It is less recommended for precision polishing be cause, precisely on account of its "bite," it is less suitable for removing infinitely small layers of material '

"Actually, the precision attainable in figuring with ceria seems to be at least equal to that attainable with rouge The diagram shows the errors of two mirrors, one polished entirely with rouge, the other started with rouge but taken most of the way with ceria and finished with twice-levigated ceria The curves of shape were determined by a method practically identical with that described by F. B. Wright in 'A.TM.' (chapter on accuracy in parabolization). The knife-edge readings were made-by Everest's shadow crest method, and the integration was done mechanically instead of numerically.

("Incidentally, these curves highlight the necessity of using a micrometer screw to move the knife-edge when a deep curve is to be measured. Although none of the discrepancies shown in the diagram would be detectable, let alone measurable, by the pencil mark and ruler technique, the mirrors are not actually exceptionally good, being only comfortably within the tolerances required for their intended applications.)

"In the upper graphs, observed knife-edge displacements (circles) are compared with the ideal values (curves). The lower graphs show the disparity between the observed and ideal cross-sections along radii from the center outward. The wavelength () in terms of which the tolerance is expressed is the wavelength of maximum visibility.

"It will be seen that the ceria-polished mirror was brought much closer to perfection than the rouge-polished one. It would have been possible to have brought either closer still, but there is no sense in further working an optical surface which is within the assigned (Wright) tolerance."

CERIUM OXIDE may be had from the Universal Shellac and Supply Co., 401 Broadway, New York 13, N. Y., which, by the way, is not the supplier mentioned, in part adversely, by Lucy and, anyway, ceria has greatly improved.

Commenting on this new polishing abrasive, the Ferson Optical Co., Biloxi, Miss., which has been using it says: "It has a heavy drag in polishing, polishes faster than most rouges, seems to have less tendency to scratch, in our experience has more tendency to turn edges but not much more, and should be used with a hard lap."

Barnesite, a somewhat expensive polishing abrasive from the same source, is held to be even better than cerium oxide by many professionals. (Incidentally, while on abrasives, the new "garnet fines" developed for wartime optical uses, and now being obtained from the same source, are rapidly moving in on fine emeries as finished abrasive for grinding.)

The Society for Amateur Scientists (SAS) is a nonprofit research and educational organization dedicated to helping people enrich their lives by following their passion to take part in scientific adventures of all kinds.