260 M. A. de Candolle on Germination 
Omitting these irregularities, the curve of each species ascends 
at first slowly, and the difference between each species is some- 
what considerable. Subsequently all the curves become approxi- 
mated and nearly parallel to the line of temperatures; and 
finally they diverge and separate towards the top. 
Hence it results that the relative order of evolution of the 
seeds is different, according to whether the low, the mean, 
. or the high temperatures are considered. The lines cross, like 
the limits of the distribution of the species in geographical botany, 
and partly from the same causes. 
It has been proposed, for the purpose of measuring the tem- 
perature required for vegetative functions, considered either in- 
dividually or in the whole of the life of a plant, to add the ther- 
mometric degrees day by day, from the commencement to the 
end, either of the function or of the life of the individual. Ac- 
cording to the calculations of M. Boussingault upon cultivated 
annual species, and those which I have published upon some 
indigenous species, nearly the same sum of degrees is found 
for the performance of one function as for all the functions of 
the same species. If the temperature has been higher, vegetation 
will have proceeded more rapidly, and vice versd, so that one of 
the numbers nearly compensates the other. As temperature 
and time are absolutely different elements in their essence, as we 
adopt them in part only of their infinite extension, and we ar- 
bitrarily divide this part into degrees and days, there is no @ 
priori reason why the days of duration should exactly compensate 
the degrees. If that happens, it is a fact, at first suspected, 
then proved; and that is all. The question is to determine to 
what point this law, which is empirical in its nature, is founded 
in reality; and, as I stated at the commencement, there is a certain 
interest in acquiring assurance by direct observations, in regard 
to a function which is less complicated than the others, and 
where heat exerts its influence without light. The calculation 
may be made in two ways—either by adding together all the 
degrees above zero, or by deducting the degrees which are useless 
to the species in the function in question, and then adding the 
other degrees, up to the moment at which the function is ac- 
complished. The latter mode appears, @ priori, more logical ; but 
the state of ignorance in which we nearly always are, in regard 
to the minima, prevents our employing it. The following are 
the numbers in relation to the species observed. I shall quote 
only three species, the remainder presenting analogous facts. 
Trifolium repens* at 5°7 requires ten days (of twenty-four 
* This species is not indicated in the plate, to prevent complication. It 
proceeds parallel with linseed in the lower degrees; afterwards, from 21° 
to 25°, it is almost identical with maize; and still higher, it separates 
from. it. 
