PROBLEMS OF HEREDITY, HORTICULTURAL INVESTIGATION. 55 
Let us recognise from the outset that as to the essential nature of 
these phenomena we still know absolutely nothing. We have no glimmer- 
ing of an idea as to what constitutes the essential process by which the 
likeness of the parent is transmitted to the offspring. We can study the 
processes of fertilisation and development in the finest detail which the 
microscope manifests to us, and we may fairly say that we have now a 
thorough grasp of the visible phenomena ; but of the nature of the 
physical basis of heredity we have no conception at all. No one has 
yet any suggestion, working hypothesis, or mental picture that has thus 
far helped in the slightest degree to penetrate beyond what we see. The 
process is as utterly mysterious to us as a flash of lightning is to a 
savage. We do not know what is the essential agent in the transmission 
of parental characters, not even whether it is a material agent or not. 
Not only is our ignorance complete, but no one has the remotest idea 
how to set to work on that part of the problem. We are in the state in 
which the students of physical science were in the period when it was 
open to anyone to believe that heat was a material substance or not, as 
he chose. 
But apart from any conception of the essential modes of transmission 
of characters, we can study the outward facts of the transmission. Here, 
if our knowledge is still very vague, we are at least beginning to see how 
we ought to go to work. Formerly naturalists were content with the 
collection of numbers of isolated instances of transmission — more 
especially, striking and peculiar cases — the sudden appearance of highly 
prepotent forms, and the like. We are now passing out of that stage. 
It is not that the interest of particular cases has in any way diminished — 
for such records will always have their value — but it has become likely 
that general expressions will be found capable of sufficiently wide appli- 
cation to be justly called "laws" of heredity. That this is so is due 
almost entirely to the work of Mr. F. Galton, to whom we are indebted 
■for the first systematic attempt to enunciate such a law. 
All laws of heredity so far propounded are of a statistical character 
.and have been obtained by statistical methods. If we consider for a 
moment what is actually meant by a " law of heredity" we shall see at 
once why these investigations must follow statistical methods. For a 
" law " of heredity is simply an attempt to declare the course of heredity 
under given conditions. But if we attempt to predicate the course of 
heredity we have to deal with conditions and groups of causes wholly 
unknown to us, whose presence we cannot recognise, and whose magni- 
tude we cannot estimate in any particular case. The course of heredity 
in particular cases therefore cannot be foreseen. 
Of the many factors which determine the degree to which a given 
character shall be present in a given individual only one is known to us, 
namely, the degree to which that character is present in the parents. It 
is common knowledge that there is not that close correspondence between 
parent and offspring which would result were this factor the only one 
operating ; but that, on the contrary, the resemblance between the two 
is only a general one. 
In dealing with phenomena of this class the study of single instances 
reveals no regularity. It is only by collection of facts in great numbers, 
