81911.] Spontaneous Cancer in Mice. 539 
while in some the growth in the spontaneously affected mice was behind that 
in the old normal mice. The differences which exist in the two first- 
mentioned experiments are small, and, keeping in view the fact that the 
conditions in spontaneously affected and in normal mice, even at the best, 
are not strictly comparable, there does not seem to be evidence of an 
essentially enhanced suitability for growth of cancer cells in general, in 
mice spontaneously affected with cancer as compared with normal mice. 
The experiments show that growth nearly always follows when a fragment 
of a spontaneous tumour is re-introduced into the animal in which the 
tumour has arisen, whereas the result is positive only in a very small per- 
centage of normal animals. A general or constitutional idiosyncrasy 
favourable to the growth of its own cancer seems to be present in the 
spontaneously attacked animal, but it only apples to cells of its own body 
and not to cancer cells in veneral. On investigating spontaneously affected 
cancer-mice from the standpoint of the soil they offer for the growth of 
tumour cells which have originated in other individuals, one finds that the 
conditions then do not differ much from those obtaining in normal animals. 
The same inference can be drawn from the results obtained when cancer- 
mice are tested as to their suitability for the growth of propagated tumours, 
and the result compared with that obtained in normal mice under conditions 
as similar as possible. The comparison demonstrates that there is no 
evidence of general constitutional changes in the spontaneously attacked 
animals, greatly favourable to the growth of cancer cells in general, which 
can explain the appearance of a cancer. The results of autologous implanta- 
tions are to be explained by the fact that the cells introduced belong to the 
same organism, where they have the full command of the supply of vessels 
and supporting tissues, and because they can be reintroduced either 
without setting up any inimical reaction, or if a reaction directed against 
them is set up, being in congenial surroundings, they are indifferent to it. 
Experiments have been made to ascertain if reactions similar to those 
observed in normal mice against propagated tumours take place in 
spontaneously attacked mice, and if they do, whether they are efficient 
against their own tumour. Cancer-mice are capable of bringing trans- 
planted tumours to absorption in the same way as normal mice do, and 
there is no reason to assume that the reactions leading to absorption differ 
essentially in spontaneously attacked mice and in normal mice, when the 
tumour cells introduced are derived from another individual. In some 
rare instances there is evidence that a process of spontaneous absorption 
or healing may also take place in a mouse towards its own spontaneous 
tumours, but it is very rare, and the evidence points rather to the 
