DECEMBER 17, 1897. ] 
_ Species. Here germ cells are not produced 
from cells not destined to that purpose as 
in the begonia leaf, but lost parts are repro- 
duced by what may be termed (and in fact 
is) an exaggerated process of healing. In 
other plants the power of reproducing lost 
parts is present in a much smaller scale, 
and only comparatively trifling injuries are 
healed ; 7. e¢.,asmall fragment cannot repro- 
duce the whole, though the whole can repro- 
‘duce lost fragments. Among animals, ow- 
ing to the greater specialization of the cells 
and the more complex condition under 
which they live, this power of reproducing 
lost parts is present in general to a much less 
extent than among plants. Low in the scale, 
as we see, a fragment of sponge, for instance, 
ean reproduce the whole. Higher in the 
scale, a starfish can reproduce a ray, a lob- 
ster a claw, a lizard its tail, and so forth, 
but none of these parts can reproduce the 
whole; that is done solely by germ cells. 
Higher yet, as among birds and mammals, 
the power of reproducing lost parts is com- 
paratively very trifling; important and com- 
plex parts cannot be restored. Wounds 
and mutilations are healed, but, if serious, 
very imperfectly, for only scar tissue re- 
places the normal tissues which were lost. 
We see, then, that the reproduction of 
lost parts, whether it be on a very great 
and perfect scale, as when a fragment re- 
produces a wholeas in asponge, or whether 
it be on a very small and imperfect scale, 
as when a wound is healed in one of the 
higher animals, is a process of the same 
order. Now, we speak of a scar in man, 
for example, as an acquired character ; but 
who would dream of speaking of all that 
which is reproduced by the fragment of a 
sponge or a begonia leaf as a character ac- 
quired by the fragment. Moreover, when 
one of the higher animals is mutilated, as 
when a dog loses his tail, we lump to- 
gether both the mutilation and the tissue 
with which the lost part is replaced (i. ¢., 
SCIENCE. 
899 
the scar) as a single acquired character. 
But, even if we should agree for conven- 
ience to regard the scar as an acquired 
character, surely the mutilation ought not 
to be so designated, but should rather be 
termed (as I venture to suggest) an enforced 
character. We see, moreover, that the 
power of reproducing lost parts to a greater 
or less extent persists throughout organic 
nature, but that this power is vastly greater 
low in the scale than higher. In other 
words, if we agree to regard such reproduc- 
tions as acquired, observation proves that 
the power of acquiring them is very much 
greater low in the scale (e. g., sponge) than 
it is higher (e. g., man). 
On the other hand, there is another class 
of acquired characters—perhaps the only class 
to which the term should properly be applied— 
the power of acquiring which is greatest 
among the highest animals, and apparently 
is little or not at all present among the 
lower animals, nor in the whole of the 
plant world. I speak of such characters as 
arise as a result of exercise and use, as, for 
instance, the increased muscular power of 
anathlete. In the plant world no charac- 
ters can, of course, be acquired as a response 
to the stimulation of exercise and use. 
Plants, therefore, of necessity, attain their 
full development in the absence of all other 
stimulation than such as is supplied by 
sufficient foodand warmth. Of such plant- 
like animals as sponges the same also, of 
necessity, is true. It is true, with possi- 
ble exceptions, even of such active ani- 
mals as insects. Thus a pupa may de- 
velop into a perfect insect while lying 
quiescent. The lower vertebrates, such as 
fish and reptiles, have also little or no power 
of developing in response to the stimulation 
of use and exercise; apparently they are 
able to grow into normal, adult animals in 
its absence; thus if a tadpole finds its way 
through a crevice into a small cavity, and 
is able to obtain sufficient food, it develops 
