530 



SCIENCE. 



[N. S. Vol. XV. No. 379. 



small enclosure of fine wire gauze in a dish 

 of water. Several males were placed in the 

 dish outside the enclosure, but none of them 

 paid the slightest attention to the females, 

 although they seized the females quickly- 

 enough when the enclosure was raised and 

 the females were allowed to scatter through 

 the dish. It is only when the males acci- 

 dentally collide with the females while 

 swimming that any attempt is made to 

 seize them. When a female collides with 

 another amphipod she curls up and remains 

 quiet for a time, when, if not seized, she 

 soon passes on. When two males collide, 

 each apparently attempts to seize the other 

 and carry him about as a female would be 

 carried. Males have the instinct to seize 

 and carry about other amphipods they meet 

 with, and are only prevented from so doing 

 by the similar attempts of the other indi- 

 vidual. Males which are mutilated by the 

 removal of the large second pair of gnatho- 

 pods, so that they are no longer able to 

 make effective resistance, are seized and 

 carried about by other males just as females 

 would be carried. Sex recognition in tliis 

 species is apparently determined by the 

 different modes in which the two sexes 

 react to the contact of other individuals. 



Some Notes on Hyhridism, Variation and 

 Irregularities in the Division of the 

 Germ-cell: Michael F. Guyer. 

 At one stage in the maturation of germ- 

 cells, preceding the true reduction division, 

 bivalent chromosomes are formed ordina- 

 rily ; that is, only half of the regular num- 

 ber of chromosomes appear, but each of the 

 new chromosomes is apparently double and 

 equivalent to two of the simple type. In 

 the spermatogenesis of hybrids, the forma- 

 tion of the bivalent chromosomes is fre- 

 quently incomplete or defective, so that the 

 resulting divisions are irregular and un- 

 equal. The greater the difference between 

 the two individuals crossed, the more 



marked is the disturbance in the matura- 

 •tion of the germ-cells of the hybrid off- 

 spring. In a paper two years ago before a 

 meeting of the Western Naturalists (ab- 

 stract. Science, February 16, 1900), I dis- 

 cussed this point in the case of hybrid pig- 

 eons and I suggested that these peculiari- 

 ties inchromosome formation might pointto 

 a tendency in the chromatin of each parent 

 species to retain its individuality, and that 

 the extreme variability seen in the offspring 

 of fertile hybrids was possibly to be attrib- 

 uted to this variability in chromatin 

 distribution. In hybrid plants (cannas) I 

 have since determined that practically the 

 same irregularities occur, and, recently, 

 Juel described abnormalities in the germ- 

 cells of hybrid plants which are in nearly 

 every respect parallel to those which I 

 found in the pigeon; hence it seems to me 

 that the same possible interpretation pre- 

 sents itself. Moreover, perhaps the same 

 conception will hold in the case of the many 

 plants, such as the geranium or apple, 

 which will not come true from seed, but 

 require propagation by means of slips or 

 grafts. 



To test this I have recently undertaken a 

 study of the formation of the pollen grains 

 in the geranium and I find that in it, as in 

 hybrids, irregularities in the first division 

 of the pollen mothei'-eell frequently occur, 

 though in a less degree. In answer to the 

 question as to why a plant will come true 

 from a graft or slip and not from seed, it 

 seems possible that we may have a clue in 

 this apparent inability of the chromosomes 

 to fuse normally to form the bivalent type 

 of chromosome. In hybrids it would seem 

 that the chromosomes from each parent lie 

 side by side and divide in an ordinary man- 

 ner to construct and maintain the body, 

 but that when the germ-cells are to be ma- 

 tured the usual doubling of chromosomes 

 which occurs at such times is incomplete, 

 the result being that the chromatin is un- 



