Pollen Development in Lactuca . 391 
the result of transverse segmentation of certain chromosomes. In this 
connexion it must be mentioned that Collins (1912) on morphological 
grounds, and as a result of crosses between maize and teosinte ( Euchlaena ), 
has developed the interesting theory that maize arose as a hybrid between 
Euchlaena and some unknown member of the Andropogoneae. This view 
is by no means proved and is still being controverted by Weatherwax and 
others. Kuwada, however, believes he has cytological evidence in its 
support, for he finds that in hybrids between Euchlaena and maize, and 
also in maize hybrids (Fig. 3 , p. 75 ), some of the bivalents "are composed of 
chromosomes of unequal length. Since the chromosomes of Euchlaena 
(10 pairs) are longer than those of various Andropogoneae, it is thought 
that the origin of the unequal pairs is to be accounted for in this way. 
The presence of eleven and twelve bivalents in some individuals of sugar 
maize is ascribed to a tendency of two chromosome pairs to undergo 
constriction and transverse segmentation, these two being descended from 
Euchlaena chromosomes. These theoretical conclusions are suggestive, but 
will require much stronger evidence before they can be accepted. 
In animal cytology several clear-cut instances of chromosome fusions 
are known. Cases in Orthoptera will be briefly considered. McClung 
(1905) described in three species of Hesperotettix a hexad multiple chromo- 
some in the heterotypic prophase formed by the temporary union of 
a particular tetrad (bivalent chromosome) with the dyad X-chromosome. 
This union never fails, and there are constant size differences in the hexad 
of the three species of Hesperotettix studied. Such a hexad has been 
observed in three Orthopteran families, Acrididae, Locustidae, and Phas- 
midae. In Hesperotettix speciosus (McClung, 1917) in the male the 
X-chromosome becomes attached to a euchromosome before synapsis, 
forming one limb of a V-shaped element with unequal arms. During 
synapsis the mate of this euchromosome pairs with it and later separates, 
but throughout maturation the X remains attached to this chromosome and 
enters the spermatid nucleus in this condition. In H. pratensis the multiple 
differs from that of H speciosus only in the proportion of its parts, while 
in the two other species no multiple (i. e. no fusion) occurs. In H viridis 
the hexad is formed, and also one or two octads may occur by the 
association of two euchromosome tetrads. The latter process appears to 
be exactly similar to what occurs in Lactuca. The X also may remain 
free. This gives rise to six types of conditions, in which the apparent 
number of chromosomes varies from nine to twelve due to different 
combinations of these fusions. But the particular associations are per- 
manent for the individual, the same type of multiple being found 
throughout all its germ cells. In the related genus Mermiria multiple 
chromosomes were lacking in two of the species investigated, but in 
M. bivittata a large hexad was formed as in Hesperotettix. In all cases 
