754 



SCIENCE 



[N. S. Vol. XXVI. No. 674 



than for the comparatively undisturbed 

 fields. 



Upon return to Washington, the investiga- 

 tion was amplified so as to embrace another 

 magnet of different style, make and magnetic 

 moment, and the weighings made for eight 

 equidistant orientations of magnet (north end 

 towards magnetic north, northeast, east, etc.). 

 Besides this, the balance itself was swung in 

 order that the beam would not always be di- 

 rected in the same way for the same orienta- 

 tion of magnet so as to esclude in every way 

 possible any effect that might be attributed 

 to magnetic impurities remaining in the bal- 

 ance, which the special tests failed to disclose. 

 It was found that the orientation of lalance 

 had no appreciable effect upon the results 

 obtained. 



However, a systematic curve of differences 

 or residuals from the mean weight, for the 

 eight orientations of magnet, resulted — some- 

 what similar to the deviation curves which 

 represent the effect on the compass of the iron 

 on board a vessel. Two waves of about the 

 same amplitude were clearly discernible — a 

 semicircular one and a quadrantal one — the 

 latter apparently associated with the inductive 

 effect of the earth on the magnet in its various 

 positions. On account of the presence of the 

 quadrantal term, reversal of magnet does not 

 eliminate the effect on the weight due to the 

 outstanding residual magnetic force. Hence 

 the mean result of weighings of a magnet in 

 two positions 180° apart will not necessarily 

 give the true weight, or say the weight which 

 the same substance would have if demagnet- 

 ized. This was proved also by repeated mag- 

 netizations and demagnetizations of two dif- 

 ferent magnets. To get the true weight of a 

 magnet within the accuracy attainable with 

 the balance used, the weighings of the sub- 

 stance when magnetized would have to be 

 made for at least eight equidistant positions. 



The observations were repeated on three 

 days, October 10, 11 and 12 at the Coast and 

 Geodetic Survey Magnetic Observatory at 

 Cheltenham, Md., and practically the same 

 results obtained as before. 



For the two stations, Washington and Chel- 

 tenham, the range in the results for the vari- 



ous orientations of magnet was about 0.05 mg. 

 (1/660,000 part of weight of magnet). 

 The investigation is being continued. 



R. L. Faris, 

 Secretary 



DISCUSSION AND CORRESPONDENCE 



VARIATION OF ENVIRONMENT 



The present writer has always been an 

 advocate of the theory that variation of or- 

 ganic beings is influenced by the environment, 

 and he does not believe in the existence of so- 

 called spontaneous or congenital variation. 



Since it is a well-established fact that no 

 two individuals of the same species are abso- 

 lutely identical, variation seems to be general, 

 and it has recently been alleged that variation, 

 the natural diversity, organisms, is a funda- 

 mental law, and is to be observed even when 

 the environment remains unchanged.' But, in 

 my opinion, this is not correct, and I believe 

 that no two individuals of any organic form 

 grow up under identical conditions." 



The idea that the latter may happen, that 

 two individuals develop under the same en- 

 vironment, demonstrates that those who hold 

 it look upon environment in a very superficial 

 way, and do not appreciate the great variety 

 of conditions involved, and thus it does not 

 seem to be amiss to call here attention to the 

 variety of features which constitute environ- 

 ment, and we shall see, if we understand this 

 properly, interesting correlations to certain 



' See Cook, in Science, September 7, 1906, p. 

 306,. " Individual diversity persists in spite of 

 uniformity of conditions." 



' In Science, July 12, 1907, p. 50, Cook quotes 

 my sentence (Science, December 7, 1906, p. 729) : 

 " if the environment remains uniform, perfect uni- 

 formity of individuals will result," but misunder- 

 stands it entirely, believing that I hold the opinion 

 that uniformity of individuals may or does ac- 

 tually exist. He would not have made this mis- 

 take, if he had paid due attention to the sentence 

 immediately following this one : " but since it is 

 practically impossible," etc. Cook's definition of 

 Amphimixis in the same article is highly sur- 

 prising to me, and I should like to know where it 

 is to be found in Weismann's writings, for I have 

 never come across it. 



