134 THE AMERICAN NATURALIST. [Vor. XXXIX. 
willows, asters, and the Juglandaceze which Prof. Sargent directs 
attention to as undoubtedly illustrating the effects of hybridiza- 
tion. Such potential species have always been a source of great 
difficulty to the systematic botanist who has been unable to 
properly define their limitations in such a way as to make them 
recognizable at all times and under all circumstances. The 
recent separation of six new species of violets from V. cucullata 
by Greene,' and the large number of species of Crataegus now 
recognized by Prof. Sargent can only be explained either as 
mutant forms or as new species according to the Mendelian law. 
Their acceptance as valid species shows that our former notions 
of the relations of hybrids and species have undergone radical 
change, and that we can no longer accept the limitations which 
were formerly supposed to be valid. If we are to accept as 
species those forms only in which there is no reversion to the 
parental type, and in which there is an absolute fixity of charac- 
ters, then the hybrid in question cannot be regarded as a species; 
but if, on the other hand, we accept as species, as is now the 
custom, those forms which possess strongly defined potentialities, 
and in which the tendency to reversion is relatively weak, then 
the hybrid in question must be held to be a valid and distinct 
species as much as any other plants so defined, and in this sense 
it should be known as C. teast, n. Sp. 
It only remains on the present occasion, to define the diag- 
nosis for the genus Catalpa, and present a differential key for 
the recognition of the species, as follows : — 
CATALPA. 
parenchyma more or less resinous, often strongly so, generally associated 
squarish and forming a limiting layer on the outer face of the growth ring 
from which irregular tracts may project radially inward, or again form a 
secondary zone. Vessels numerous, radially or sometimes tangentially 
compounded, with or without thyloses. 
Radial— Medullary rays somewhat resinous ; the cells often higher than 
1 Ottawa Nat., vol. 12, pir: 
