8yu 



SCIENCE. 



[N. S. Vol. XX. No. 521. 



as to remove all danger of accidental fertiliza- 

 tion. The eggs were then allowed to lie in 

 sterilized sea-water for an hour and a half, 

 during which time they showed no sign of 

 having been fertilized. Individual eggs were 

 then cut horizontally, one by one, into an 

 upper nucleated fragment containing the 

 maturation figure, and hence the two centro- 

 somes, and a lower non-nucleated fragment. 

 The latter was subjected to a solution of cal- 

 cium chloride in sterilized sea-water. After an 

 hour they were replaced in ordinary sterilized 

 sea-water. As a result of this experiment, 

 many, indeed almost all of the non-nucleated 

 fragments produced asters, sometimes single, 

 sometimes in large numbers (in one case more 

 than a score of asters were observed in a single 

 fragment). Many, practically all, asters con- 

 tained centrioles. No cytasters developed in 

 the control eggs allowed to remain in sterilized 

 sea-water. Sections of the non-nucleated 

 fragments thus treated showed that the asters 

 and centrioles are identical in structure with 

 those of an entire egg subjected to a solution 

 of calcium chloride, while preparations of the 

 corresponding nucleated half demonstrated 

 the presence of the two original centrosomes. 

 No other conclusion, therefore, is possible 

 than that the centrioles of the non-nucleated 

 half have been formed de novo. The experi- 

 ment, I think, verifies the conclusion reached 

 in Wilson's experiment, and is contrary to 

 the negative result recently published by Pe- 

 trunkewitsch. A detailed presentation of the 

 evidence will be given hereafter. 



N. Tatsu. 



EARLIEST NOTICE OP AMERICAN PROBOSCIDEA. 



The opinion is current and appears to be 

 well founded that vertebrate paleontology in 

 this country had its beginning in Thomas 

 Jefferson's description of ' mammoth ' re- 

 mains from Virginia in 1787,* and of the 

 bones of Megalonyx a dozen years later. 



So far as scientific investigation goes, this is 

 undoubtedly true, yet it is interesting to recall 

 that fossil elephant remains have been known 

 from the western world for a much longer 



*' Notes on the State of Virginia' (London, 

 1787). 



period, and from Europe (Sicily) since at 

 least the days of Empedocles of Agrigentum. 



Not only was Columbus particularly en- 

 joined by the Spanish sovereigns to bring 

 back with him from America ail manner of 

 natural products, but in later years Hernan- 

 dez, private physician to Philip II., and other 

 distinguished functionaries were sent to 

 Mexico for the special purpose of reporting 

 upon the vegetable and zoological curiosities 

 of the country. It was by these travelers, 

 amongst the most prominent of whom besides 

 the afore-mentioned were Oviedo, Acosta and 

 Garcilaso, that fossil proboscidean remains 

 were collected on the elevated plateaux of 

 Mexico, Peru and elsewhere. 



Detailed references are given in the second 

 volumes respectively of Cuvier's ' Ossemens 

 Fossiles ' and Humboldt's ' Cosmos ' to vari- 

 ous old Spanish works in which these fossils 

 were described as belonging to a race of 

 human giants, the localities furnishing them 

 being called ' Campos de Giganies.' The ab- 

 surd discussions of ' Teutohochus rex ' in the 

 early part of the seventeenth century are of 

 interest only for revealing the crude state of 

 natural science at that period. C. R. E. 



CURRENT NOTES ON METEOROLOGY. 



TEMPERATURE EST CYCLONES AND ANTICYCLONES. 



At the 1904 meeting of the British Associa- 

 tion, Mr. A. Lawrence Rotch summarized the 

 results of observations obtained at Blue Hill 

 Observatory during Siiiite flights, at different 

 seasons, in areas of high and low pressure, up 

 to about 12,000 feet. The mean decrease of 

 temperature, computed by stages of 1,600 feet, 

 is nearly constant, averaging 1° P. in 376 feet 

 of ascent. Whether the whole column of air 

 in a cyclone is warmer than the correspondirig 

 air in an anticyclone (as the convectional 

 theory requires) depends chiefly upon whether 

 its initial temperature at the ground is higher 

 than that of the anticyclone, which is usually 

 the case. If the data obtained from kite 

 flights on consecutive days be plotted for the 

 same height, as was first done at Blue Hill in 

 1899, it is seen that up to the height of 12,000 

 feet it is generally warmer at all levels over 



