June 19, 1908] 



SCIENCE 



947 



with the pharyngeal wall, on one hand and 

 with the pouch-like anlage of the thymus, 

 on the other. Both connections are then 

 lost and the body disappears before the end 

 of the seventh day. This epithelial body 

 has not been observed on the right side in 

 any of the series studied. 



In connection with the second gill-cleft 

 a cell-thickening in every way similar to 

 the thymus anlage of the third gill-cleft is 

 formed simultaneously with the latter, but 

 it begins to degenerate about the sixth day 

 and takes no part in the formation of tho 

 thymus. 



The Chromosomes in the GrylUdce: W. J. 



Baumgaktner, University of Kansas. 



In a former paper the writer showed that 

 the shape assumed by the chromosomes of 

 Gryllus domesticus during the prophase 

 and metaphase of the first spermatocyte 

 division was constant and could be used as 

 a differential characteristic. The sperma- 

 tocyte has 11 chromosomes. 



The examination of nine other species of 

 this family shows a great variation in 

 shape, size and numbers of chromosomes. 

 The highest number is 29 spermatogonic 

 elements and the lowest 13. The accessory 

 in all the species studied exhibits the same 

 peculiarities of shape and behavior as de- 

 scribed for G. domesticus. 



Three species have 23 chromosomes agree- 

 ing with the Acrididse as described by Mc- 

 Clung and others. 



In several species certain chromosomes 

 can be readily recognized by their shape 

 and size. The tree crickets have 19 

 spermatogonic chromosomes, of which the 

 accessory and four others are very long. 

 In the prophase the four long ones unite 

 into two enormous rings. If these rings 

 are multiples, i. e., quadruple, as one might 

 easily conclude, then the number of ele- 

 ments in the two species examined is really 

 23. 



Apitlus agitabor has 7 spermatocytic 

 chromosomes. Of these one is the acces- 

 sory, one is ordinary in size and simple in 

 shape, and the other five are large rings. 

 Should these last prove to be quadruple 

 then the real unreduced number here is 

 also 23. 



Nemobius fasciatus has 8 chromosomes 

 in the first spermatocyte, of which 4 are 

 much larger and ring-shaped. Quad- 

 rupling these gives a real number of 23. 

 But Nemobius exigeus has nine elements, 

 of which also 4 are large rings. Quad- 

 rupling these and doubling all but the 

 accessory, we get 25. The number of 

 elements in two other species can not be 

 reduced to 23 by any marked size and 

 shape differences. 



Our result shows the three species have 

 23 chromosomes. Four species can be re- 

 duced to 23 by using the "multiple chro- 

 mosome" idea and three species can not. 



The Gryllidffi also have much greater dif- 

 ference in shape and size of chromosome 

 than the Acrididas. The difference of 

 shape has a marked tendency to be con- 

 stant. 



Was the Hydroid or the Medusa the Orig- 

 inal Form of Hydromedusce? C. C. Nut- 

 ting, University of Iowa. 



The development of typical forms of 

 Hydromedusse was briefly described and 

 the alternation of generation discussed. 



The elimination of successive terms in 

 the life histories was described and the 

 various methods of reproduction presented, 

 showing the two lines of divergence to the 

 medusa-form, on the one hand, and the hy- 

 droid form, on the other ; ending in the free 

 medusa without fixed hydranth and the 

 fixed hydranth without medussB, respect- 

 ively. 



The argument of ontogeny as showing 

 the phylogeny of the group points to the 



