May, 1923] 
SHULL — SPECIES CONCEPT 
225 
distinct from the species common to eastern America and Europe. In 
other words, it is convenient to differentiate between groups of organisms 
occupying different geographical regions, even though the differentiating 
characteristics are few and of relatively insignificant magnitude, while it is 
impracticable to separate intermingled groups differentiated by similar 
relatively slight characters. 
A second kind of sterility is exhibited within the very distinct large- 
flowered species, Bursa grandiflora Bois., native of the regions about the 
Aegean Sea. In this species I have been able to demonstrate that there 
are two groups of plants, A and B, not visibly differentiable, both of which 
are sterile or nearly so when selfed or when crossed with other individuals 
of their own kind, but which are fully fertile when any two individuals 
mated together belong to different groups. In other words, A X self or 
A X A, and B X self or B X B, yield no offspring or almost none, while 
A X B or B X A is fully fertile and produces seeds abundantly. Here we 
are dealing with so-called self-sterility and cross-fertility; but, as already 
stated, there is also cross-sterility here, since A crossed with any other A 
plant or B with any other B plant produces no seeds. No one can maintain 
that incompatible crosses in such a case as this indicate specific distinctness 
between the parents, for it is just the most closely related individuals , those 
of like genotypic constitution, that are incompatible. 
When we consider the results of crossing together such very distinct 
species as Bursa hursa-pastoris, B. grandiflora , and B. Viguieri, an even 
more striking limitation on hybrid sterility as an indication of specific 
distinctness is impressed. The wide differences in the three species here 
named may be noted in table 1, in which are entered their several con¬ 
trasting characters. 
Table i 
Grandiflora 
Bursa-pastoris 
Viguieri 
Rosettes 
Lax 
Rather lax 
Dense 
Leaves 
Always well lobed 
Variously lobed 
But little lobed 
Surface 
Smooth, shining 
Moderately smooth 
Rugose, dull 
Stellate hairs 
Numerous 
Abundant 
Rare, small 
Stems 
Normal 
Normal 
Fasciated 
Flowers 
Large 
Medium size 
Small 
Odor 
Strong balsam-like 
Absent 
Absent 
Breeding 
Self-sterile 
Self-fertile 
Self-fertile 
Carpels 
2 
2 
3 to 8 (mode 4) 
During the past six years I have made more than twenty hybridizations 
between B. Viguieri and B. hursa-pastoris, using in each case a different 
bio type of the latter species. Hybrids have always been easily secured 
from these crosses, the fecundity being at least approximately as great as 
when pollen from the same species has been used. The hybrids have 
always been vigorous plants of the hursa-pastoris type, no matter which 
