76 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 
HYBRIDISATION AND ITS FAILURES. 
By the Rev. Professor G. Henstow, M.A., F.L.S., V.M.H., &e. 
INTRODUCTION ; DEFINITION OF A SpEcrES.—In endeavouring to 
find some clue to the interpretation of hybrids, as to why some species 
when crossed fail, while others succeed and bear fertile offspring, it is 
desirable to consider what is the present idea of a species. Two con- 
siderations were formerly maintained, viz. morphological structures and 
a presumable physiological affinity. Thus Bentham defined a species as 
follows: “A species comprises all the individual plants which resemble 
each other sufficiently to make us conclude that they are all, or may 
have been all, descended from a common parent.’’ * 
This definition may be sufficient as long as no physiological question 
is raised as to the capabilities of different species of the same genus 
intercrossing. Dean Herbert, however, soon found that another element 
must be considered, and that was interbreeding. Since the practice of 
hybridising plants has been extensively pursued ever since he wrote, the 
idea has been maintained that if two species would cross and produce 
fertile offspring, then they must be regarded as of common parentage, 
and as being only varieties of one and the same species. Thus Dean 
Herbert writes, referring to experiments of Knight: “The President 
adopted in his writings a principle or dogma, which seemed to be then 
much relied upon by botanists, that the production of a fertile cross was 
proof direct that the two parents were of the same species, and he 
assumed as a consequence that a sterile offspring was nearly conclusive 
evidence that they were of different species.’’ Hethen further adds: “I held 
also... that the production of any intermixture amongst vegetables, 
whether fertile or not, gave reason to suspect that the parents were 
descended from one common stock and showed that they were referable 
to one genus; but that there was no substantial and natural difference 
between what botanists had called species and what they had termed 
varieties. . . . If two species are to be united in a scientific arrange- 
ment on account of a fertile issue, the botanist must give up his specific 
distinctions generally and entrench himself within the genera.’’t 
Testing the question as to the more or less agreement in external 
features between so-called closely allied species being correlated with 
fertility in their hybrids, we now know that the general rule may be 
formulated that such is the case; yet there are so many exceptions that 
the suggestion of Herbert for systematists to follow must be disregarded, 
and that they must continue to describe new species and genera solely 
by the morphological characters they present. 
This is practically what is always done; so that for purely systematic 
purposes it would seem that physiological affinity must be neglected 
altogether, as, ¢.g., when masses of dried plants are sent to Kew from 
some newly explored country. 
* Introduction to the Handbook of the British Flora, 1865, p. xxxvii. 
+ Amaryllidacee, p. 336. 
