474 Fraser and Weis ford. — Further Contributions to the 
Apart from the premature longitudinal fission of the heterotype, and 
the consequent occurrence of the homotype division, the difference between 
the two methods of reduction seems to lie in the extra contraction of 
meiosis. In all probability this contraction is the second, since Moore and 
Embleton ( 21 ) have observed the formation of the gemini in the cockroach 
before a spireme appears, and Harper ( 19 ) has found synapsis in Phyllactinia 
where the chromosomes are already paired. These facts suggest that the 
first meiotic contraction rather than the second is connected with the union 
of the chromosomes. 
It seems, then, that meiosis is distinguished from brachymeiosis by its 
synaptic phase, and it seems not unlikely that this indicates the moment of 
some interchange of material between the already paired allelomorphs. If 
this be the case, the opportunity for interchange must vary considerably 
according to the extent of the synaptic contraction ; it would be of interest 
to ascertain whether there is any relation between this difference and the 
occurrence of mutation. 
We are inclined to believe that brachymeiosis, since it lacks a second 
contraction, admits of less variation in its products than meiosis, and implies 
either the separation of the entire nuclei which fused, or at any rate a 
sorting of imciltered chromosomes. Similarly it might be possible to 
regard asexual fusion as essentially a temporary expedient, the result 
of such casual conditions as proximity (as when two nuclei associated 
in the same cell fuse) ; and sexual fusion, even when reduced, as a more 
permanent and significant process implying an interchange of parental 
material. 
We are not prepared to suggest that the forms of reduction are never 
interchangeable ; but, while its phylogenetic history is perhaps the ultimate 
test of the nature of a fusion, it is noteworthy that there is no case known 
of a sexual process, however simplified, followed by any but a meiotic 
reduction. 
Spore-formation. 
Spore-formation in the Ascomycetes was first studied by Harper ( 15 ) 
in 1895, and his account has been several times confirmed. He concludes 
that the spore is bounded by the fibres of the polar aster, which bend round 
and fuse laterally to form a membrane. 
Faull ( 11 ), in 1905, described the spore as cut out by the gradual 
differentiation, from the centrosome downwards, of a limiting layer of 
‘ hyaline or finely granular ’ cytoplasm. This account, unlike Harpers, 
has been correlated with the processes observed in certain Phycomycetes. 
In 1908 it was suggested by one of us ( 13 ) that the spore is in fact 
delimited by the astral rays, but that these represent currents flowing 
