366 Gates and Rees . — A Cytological Study of 
The material originally collected in 1919 came from a sowing of com- 
mercial seed of Sutton’s ‘Dwarf Perfection’. This is essentially a Cos 
lettuce and habitually throws 1 to 2 per cent, of ‘ rogues the rogue 
resembling a Cabbage lettuce. The type, however, when carefully selfed, 
we are informed by Dr. Bateson, breeds true as far as present experience 
goes, some five hundred offspring all remaining true to type. But the 
rogues ‘ throw a profusion of forms, ranging from a small percentage of good 
Cabbage lettuces, through miscellaneous mongrels, to an occasional plant 
approaching Cos, but I doubt if a true Cos, or even a true Dwarf Perfection, 
came in some hundreds of them ’ (Dr. Bateson in letter). There are 
difficulties with the view that the rogues are cross-bred, and these prelimi- 
nary results of Bateson are referred to here as indicating all that is at present 
known concerning the relationship of the type and the rogue. Cyto- 
logically, the type and the rogue are very similar, and no certain differences 
have been established between them. 
The most satisfactory fixing fluid was found to be 1 per cent, chromo- 
acetic acid solution, and Heidenhain’s iron-alum-haematoxylin gave the 
most sharply defined and satisfactory staining results for most purposes. 
The sections were cut chiefly 10 [i in thickness. The figures are all of 
cultivated lettuce, and no distinction is made between the type and the 
rogue except in the explanation of plates at the end of the paper. 
Early History of Pollen Mother-cells and Tapetum. 
In a longitudinal section of the buds of lettuce, a very small number of 
pollen mother-cells appears in each loculus. Numerous counts in all 
the species indicate the presence of from fifteen to twenty mother-cells 
in each loculus, from which usually about sixty pollen grains develop. 
Moreover, a striking feature of all three species is the loose arrangement of 
the cells in even the earliest pollen mother-cell stage. The separation 
of the pells begins in the archesporium. As shown in Figs. 1—4, not only 
are the pollen mother-cells frequently separated from each other even 
before synapsis, but they usually lie free from the tapetal cells, and the latter 
are for the most part more or less completely separated from each other. 
This last condition is so different from the usual condition in flowering 
plants in which the tapetum forms a compact and continuous layer of cells, 
that it was at first thought to be possibly a result of long cultivation. But 
comparison with the two wild species showed this condition to be equally 
characteristic of them, and so probably of the genus as a whole. 
Another peculiar feature which is associated with the loose arrangement 
of the tapetal cells is the fact that a variety of transitional stages occur 
between tapetal cells and pollen mother-cells. During the period when the 
mother-cells are in synapsis and the postsynaptic spireme stages, the tapetal 
