multitude of plants, show very clearly that the crossed plants exceed
the self-fertilised in height, but they do not tell by inference
anything about their respective mean values. If it should happen that a
series is known to follow the law of error or any other law, and if the
number of individuals in the series is known, it would be always
possible to reconstruct the whole series when a fragment of it has been
given. But I find no such method to be applicable in the present case.
The doubt as to the number of plants in each row is of minor importance;
the real difficulty lies in our ignorance of the precise law followed by
the series. The experience of the plants in pots does not help us to
determine that law, because the observations of such plants are too few
to enable us to lay down more than the middle terms of the series to
which they belong with any sort of accuracy, whereas the cases we are
now considering refer to one of its extremities. There are other special
difficulties which need not be gone into, as the one already mentioned
is a complete bar."]

Mr. Galton sent me at the same time graphical representations which he
had made of the measurements, and they evidently form fairly regular
curves. He appends the words "very good" to those of Zea and Limnanthes.
He also calculated the average height of the crossed and self-fertilised
plants in the seven tables by a more correct method than that followed
by me, namely, by including the heights, as estimated in accordance with
statistical rules, of a few plants which died before they were measured;
whereas I merely added up the heights of the survivors, and divided the
sum by their number. The difference in our results is in one way highly
satisfactory, for the average heights of the self-fertilised plants, as
deduced by Mr. Galton, is less than mine in all the cases excepting one,
in which our averages are the same; and this shows that I have by no
means exaggerated the superiority of the crossed over the