952 THE AMERICAN NATURALIST. [Vor. XXXV. 
into the new form. Another example of the same process, 
first made known by Bickford, is that of Tubularia. In this 
form it is even more evident that a piece of the stem may be 
transformed into a new hydranth without new tissue forming 
beyond the level of the cut-end. 
In planarians we find not only new tissue developing at the 
cut-end, but also a transformation taking place in the old parts. 
The Protozoa also regenerate by changing over the old part 
into a new form. From the figures given by Gruber and 
others there can be little question that the piece is transformed 
directly into a new whole, although this important fact has not 
been emphasized by those who have written on the subject. 
A piece of Stentor, for example, if it contains one or more 
nodes of the macronucleus, produces a new Stentor having the 
characteristic form, but on a smaller scale. I have repeated 
this experiment on Stentor, and have found that small pieces 
produce new organs that are of proportionate size. 
In these cases of morphallaxis, just described, the small 
pieces change over into the characteristic form of the species, 
ie, they are new wholes of smaller size. They may subse- 
quently grow to the full normal size, and even produce sexual 
organs. On the other hand, there is known at least one case 
in which very small pieces produce not a whole form, but only 
a part of a whole. It has been shown that very short pieces 
of the stem of Tubularia may produce only the distal end of the 
hydranth, vzz.,a proboscis. Other experiments show that, never- 
theless, the material out of which the proboscis develops is 
totipotent. This problem will be more fully discussed later in 
another connection. The result also suggests direct compari- 
son with those cases in which only a part of what has been 
removed is regenerated out of the new tissue that appears at 
the cut-end, as when only a few anterior segments regenerate 
in the earthworm after the removal of a much greater number. 
In fact, I think it may be shown that there is something more 
than a mere similarity between the two cases, and that they 
both result from the action of the same factors. 
We may next examine some of the cases in which a part of 
the egg, or of the embryo, produces an entire organism. The 
