540 



SCIENCE, 



[Vol. IV., No. 



vision is made for the preservation of such 

 perishable collections as dried specimens of in- 

 sects. Under the present condition of things, 

 it is actually unfortunate for the future of 

 this science, when an enthusiast arises in some 

 local museum whose care for and interest in 

 these objects result in the accumulation of a 

 considerable collection, often containing valu- 

 able t3'pes. At his death or removal, or pos- 

 sibly the failure to retain his early ardor, the 

 chances are ten to one that the collection will 

 be ultimately destroyed. Even our best en- 

 dowed institutions have failed to make any 

 proper provision for the preservation of their 

 collections of insects and stuffed animals, — the 

 two departments of a natural-history museum 

 which require eternal vigilance. 



There are many valuable entomological col- 

 lections in the hands of specialists in this 

 country, which would find their way by gift, 

 or jby sale on easy terms, to the National 

 museum at Washington, were any reasonable 

 inducement held out to them. These collec- 

 tions contain material especially valuable for 

 the future of descriptive entomology in this 

 country. Within a few years many such col- 

 lections have been sold, either to other private 

 collectors, or perhaps to parties out of the coun- 

 try, to find their place in European museums, 

 where they are insured perpetual care. It is 

 only within three } T ears that there has been 

 even a nominal curator in charge of the collec- 

 tion of insects at the National museum ; and 

 the paltiy collection of the department of agri- 

 culture was all the authorities at the national 

 capital had to show for an entire department 

 of natural history, and one abounding in its 

 wealth of varied forms. The present curator 

 has but an honorary office, and is without funds 

 for the support of an assistant. Until pro- 

 vision is made for the proper conduct of this 

 immense department of natural history at the 

 national capital, the appointment of an honor- 

 ary curator is worse than useless. It only 

 deceives those who know no better, into the 

 supposition that collections sent to the muse- 

 um are insured proper care. They are not. 



LETTERS TO THE EDITOR. 



**# Correspondents are requested to be as brief as possible. The 

 writer's name is in all cases required as proof of good faith. 



Verification of predictions. 



The vulnerable point about Mr. Doolittle's meas- 

 ure of success (given under ' Proceedings of societies ' 

 in this number of Science) seems to me to be his com- 

 bination of the two differences of probabilities, — 



c p — c n c o — c 



o 8- 



c c 



- and -• 

 o p 



— p 



It appears clear to me that either of these differences 

 may be taken alone, with perfect propriety, as the 

 true measure, according as our concern is to test oc- 

 currences for successful prediction, or to test predic- 

 tions for fulfilment. If we allow an importance n 

 to the former test (limits of n, and 1), so that an ad 

 valorem change of 6 in this measure produces an 

 ad valorem change of nd in i, and similarly an im- 

 portance 1 — n to the latter test, these two quantities 

 will enter as exponents, and 



/c p — c\ n /c o — cV 

 \o s — o) \p s — p) 



In my opinion, the value of i is not discoverable 

 unless the value of n is given ; and this is a subjective 

 quantity. Assuming n = i, we have for i an expres- 

 sion equal to the square root of that given by Mr. 

 Doolittle, and without the fault of giving no negative 

 values to answer to perverse predictions. 



Heney Farquhar. 



The microscope for class-room demonstration. 



The following adaptation of the use of the micro- 

 scope as a sort of magic-lantern for class demonstra- 

 tion has been found so extremely useful, cheap, and 

 practical, that it is illustrated here. 



A large common kerosene 'duplex' lamp is the 

 illuminator. Superfluous light is cut off by a piece 

 of six-inch stove-pipe, which fits over the lamp-chim- 

 ney, and rests upon a horizontal collar, C, of stove- 

 pipe metal. The collar prevents the pipe from 

 shutting down too far upon the lamp, which would 

 cause the kerosene to become dangerously hot. The 

 lamp is filled at i^with a curved glass funnel; and the 

 two flat wicks, an inch and a half broad, are turned 

 by their separate keys outside of the pipe. The pipe 

 has two elbows, which conduct heat and smoke away, 

 and completely cut off the light from the top of the 

 flame. These elbows may be rotated into any con- 

 venient position. Opposite the lamp-chimney a third 

 short elbow, E, is inserted, closed by a movable cap. 

 Through this elbow the chimney can be removed, the 

 wicks trimmed, and a concave glass or tin reflector, 

 M, four inches and a half in diameter, may be placed 

 behind the flame. The flat of the wicks should be 

 parallel to this mirror. Opposite the mirror, and 

 directly in front of the flame, a plano-convex lens,- X, 

 two inches in diameter, is inserted in a hole in the 

 pipe. The light reflected from the mirror, M, passes 

 through this lens, and falls upon the reflector of the 

 microscope, whence it is made to illuminate the ob- 

 ject upon the glass slide in the ordinary way. The 

 object is magnified by a one-fifth inch or one-half inch 

 objective; the eye-piece of the microscope is removed; 

 and the image is projected upon aground-glass screen, 

 G, a foot and a half square, which is placed from 

 one to four feet in front of the microscope. The 

 screen is supported by a perpendicular iron rod and 

 cork-lined clamp, such as is in use in every chemical. 



