468 BOTANICAL GAZETTE [JUNE 
plate with their angles turned toward the interior of the spindle, and it 
is often evident at this stage that the arms of a V are not fused, but are 
simply in contact at the angle. Sometimes more than twelve V’s are 
to be observed. If, as I suspect, this fact is due to the persistence of 
the form of the occasional V-shaped granddaughter chromosomes of 
the previous division, we should expect to find now and then a V-shaped 
daughter chromosome in the metaphases of the homoeotypic division ; 
but I have not as yet observed such figures. ‘he separation in the 
metaphases is usually, at least, at the angle of the V, resulting in rod- 
shaped daughter chromosomes which are identical with the grand- 
daughter chromosomes of the heterotypic division. 
If my description be correct of the origin of the single spirem by a 
fusion of two originally separate threads, an explanation is offered for 
the sudden appearance in the heterotypic division of the reduced num- 
ber of chromosomes. Each nucleus of the plant or animal, in preparing 
for division, forms a spirem, composed of substances derived in equal 
proportions from the male and female parents. There is no fusion of 
these hereditary substances thoughout the life history of the individual, 
until the initiation of nuclear division in the spore mother-cells in the 
case of plants, or in the primary spermatocyte or the primary oocyte in 
the case of animals. In the prophases of this division, the two portions 
of the spirem, each derived ultimately from one of the parent germ 
cells, become applied to each other and fuse into a single thread, so 
effecting the mixture or interaction of hereditary qualities which was 
provided for by the fusion of the sex cells at the beginning of the 
life-cycle. ; 
As I have said, the chromomeres fuse in pairs; but just what this 
fusion involves as regards the smaller units which makeup the chromo- 
meres is a problem which at present perhaps cannot be solved by direct 
observation. If we adopt the hypothesis, frequently advanced, that the 
transmission of hereditary qualities is a function of the chromomeres, 
or of their component units, a very interesting parallel may be traced 
between the observed facts above described and the results of recent 
experimental studies of hybridization. A discussion of this and of 
related questions will be deferred until the appearance of the complete 
paper. 
It is interesting to note that De Vries (’03) has recently concluded, 
upon hypothetical grounds, that before their separation in the hetero- 
typic division the chromosomes lie side by side in pairs, each pair con- 
sisting of a paternal and a maternal segment; and that, in this condition 
