284 
\ 
ed vigor; while those brought from California here are equally affected 
with others. 
Another important fact, demonstrable by experiment, is, that when 
you produce a family of seedlings, they not only resemble the parent 
variety in all their leading qualities, but actually exhibit less hardiness 
in proportion to the amount of tuberous elaboration than the original 
plants. 
Still another fact, characteristic of these experiments, and shaking the 
foundations of the wearing-out hypothesis, is, that in all their efforts at 
reproduction de novo , the experimenters have realized no such thing as 
invariability of result. The prevailing dissimilarities are such as to be 
wholly irreconcilable with each other, and phenomena are constantly oc¬ 
curring which are inexplicable upon any theory yet started. 
Another, and still more embarrassing fact, is, that many non-tuber- 
ous plants allied to the night-shade family have taken the disease; and 
even some not so allied have showed more or less indications of it. 
Notwithstanding the difficulty of harmonizing these facts, the theory 
of exhausted energy still continues to be the favorite one adopted; and the 
prevailing remedy suggested, is, a restoration by new varieties. The main 
corollary drawn from the hypothesis is, that improvability must follow 
reproduction. In order to determine the probabilities of success in any 
given reproduction, we start with the algebraic method of assuming the 
unknown quantities, and after explaining the several facts by the very 
original process of supposing them to happen in obedience to some gen¬ 
eral law of development, we finally return to the identical proposition upon 
which we started, and record the important discovery that x=x. The 
sum of all our experiments is, that we shall succeed in ‘ proportion to 
the elevated point from which we start.’ We must start with hardy 
varieties to get hardy varieties; for there is always ‘a tendency in like 
to beget like.’ And if we do not succeed in the first reproduction, we must 
try a second, and third, and so on; practically involving our identical 
propositions, till our algebraic process returns to the original formula of 
Mike equals like.’ This is certainly a very innocent and harmless, not 
to say satisfactory, way of getitng at a result. As long as the produc- 
i 
