380 Tricotylous Races. 
the point of departure for the race ; they offer the prospect 
of providing a half race. In the third case we may ex- 
pect to obtain an intermediate race rich in tricotyls. 
Almost every year I have made experiments of this 
kind, but I was especially engaged with them in the 
spring of 1895. At that time I sowed about 20 grams of 
each of 40 species of annual plants, or in the case of 
very small seeds a somewhat smaller quantity, so as to 
investigate several thousand seedlings of each kind. I 
shall now give a list of the species falling into the third 
category, species, that is, which gave so large a number 
of aberrant forms as to justify the expectation of an 
intermediate race an expectation which has, as a rule, 
been fulfilled, as we shall see in the following section 
(5). 
TRICOTYLS FROM BOUGHT SEED. 
(Spring, 1895.) 
SPECIES DICOT- HEMI- TRICOT- TETRA- 
YLS TRICOTYLS YLS COTYLS 
Chrysanthemum inodoruin 
plenissimum .... 1000 3 32 
Silene orientalis alba . . 3000 3 70 
Papaver Rhoeas fl. pleno . 3000 1 15 1 
Clarkia pulchella alba . 4000 5 50 
Glaucium luteum . . . 16000 15 
Nigella hispanica alba . 10000 15 
Phacelia tanacetifolia . . 16000 8 18 
Helichrysum bracteatuni . 35000 9 16 3 
With the exception of Silene, Glaucium and Nigella, 
I have raised intermediate races from all these sowings. 
As I have already mentioned, hemi-tricotyls and tetra- 
cotyls are seen to be rarer than the typical tricotyls. A 
sample of seed of Lobelia Erinns, grown in the spring 
of 1902, had a very high proportion of tricotyls, viz., 
31 in 100 seedlings. 
