Reduction in Animals and Plants. 437 
malities with a tendency towards the formation of tripolar 
spindles. 
Now there cannot exist a shadow of a doubt that the great 
heterotype division of Amphibia corresponds to the great 
heterotype in Mammalia, Elasmobranchs, and Birds, any more 
than it is possible to doubt that the synaptic change corresponds 
among varieties of plants. 
But the number of cellular generations which follow the 
synapsis and the heterotype, appears by no means constant, 
either among animals or plants. In some mammals which 
I have examined, the great heterotype division is the last in 
the whole spermatogenetic series, as indeed the researches of 
Brown, Ebner, Fiirst, and Hermann had already unconsciously 
sufficed to show ; and it consequently follows that the par- 
ticular part of the spermatogenesis, in which, according to 
vom Rath, the ‘ Reductionstheilung ’ exists in Salamander, 
may be altogether wanting among mammals. In mammals 
as in Elasmobranchs the synaptic reduction does really appear 
to accomplish all that is done in that direction. But in 
Elasmobranchs two generations and one division follow the 
heterotype, and the last division is a heterotype also, with the 
same number of chromosomes as the first. I do not see how 
in the face of these facts the universality of the ‘ Reductions- 
theilung ’ in animals is to be maintained. 
Turning to the case of plants, Dr. Haecker is evidently 
himself impressed with the fact that the peculiar division 
occurring at the end of the vegetable synapsis is superficially 
similar to the heterotype, accompanying the corresponding 
stage in the development of animal reproductive cells, for he 
says on page 100 that a numerical reduction is known some- 
times to accompany the heterotype division of animals, and 
then goes on to imply that the synapsis, or reduction in rest, 
hitherto described in plants, may be followed by divisions 
having the nature of a true ‘ Reductionstheilung.’ All those 
botanists, however, who have paid the most attention to this 
subject, universally declare that no such divisions exist, and 
we are left with no other alternative than to conclude that 
