16 INVERTEBRATA CHAP. 



We thus reacli the conception that male and female chromosomes 

 remain side by side without fusing in the nuclei of the offspring 

 during all its life, but that when this organism in turn produces germ 

 cells these two kinds of chromosomes are segregated into different 

 gametes. Now this conclusion appears at first sight to accord exactly 

 with the theory to which the followers of Mendel have been led, and 

 it entirely destroys the second half of Weismann's theory of the pro- 

 duction of variations at the reducing division, by the casting out of 

 half the chromosomes, selected at random. 



This brilliant school of " Mendelians," whose labours have been 

 summarized by their most brilliant member, Bateson (1909), have 

 been led to conclude that when two strains of animals are crossed, 

 the hybrid produces two kinds of spermatozoa, or ova, as the case 

 may be, one carrying the characters of the male and the other of the 

 female with respect to each differentiating character. 



We should be wanting in our duty, however, if we allowed our 

 readers to imagine that Montgomery's theory had been fully estab- 

 lished. It is, on the contrary, only in the stage of a working hypothesis, 

 and it labours under many difficulties. Thus, its superficial agreement 

 with the Mendelian theory disappears under a deeper analysis. On 

 the idea that male and female chromosomes are distributed to distinct 

 gametes, each zygote should produce only two kinds of gametes 

 (leaving out of sight the sex-chromosomes for the present), one with 

 the maternal, one with the paternal characteristics. But the 

 Mendelian hypothesis demands two kinds of gametes with regard to 

 each differentiating character. Thus if a pea-plant have round and 

 green seeds, and if it be crossed with another having yellow and 

 angular seeds, we must expect the hybrid plants to produce seeds 

 which give rise to plants bearing yellow and green seeds, and round 

 and angular seeds ; but all the yellow seeds will not be angular nor 

 will all the green be round ; on the contrary there will be four 

 categories of seeds produced, viz. : yellow round, yellow angular, 

 green round and green angular. It is not easy to see how Mont- 

 gomery's hypothesis can be fitted in with this breaking up of the 

 parental hereditary potencies into factors which are distributed 

 among the germ cells independently of one another. Agar has, 

 however, pointed out that in the stage of zygonema, when the first 

 pairing takes place between chromosomes, there is opportunity for 

 the exchange of material between them, and that when they again 

 separate in the stage of strepsinema they may be different from what 

 they were before the pairing took place. 1 



Montgomery's theory demands, further, the belief that the identity 

 of each chromosome remains unimpaired during the resting period of 

 the nucleus, when no trace of distinct chromosomes can be detected. 



1 It has also been suggested that, previous to the reducing division, the bivalent 

 chromosomes, each of which (ex hypothesi) consists of a paternal and of a maternal chro- 

 mosome joined end to end, may not all be arranged so that their homologous ends point 

 in the same direction. If this were so, one gamete might receive one chromosome from 

 one parent and one from another. 



