Experiments in Grafting 301 
Crampton has studied the effects of uniting parts of different 
species of moths. The pupe of the species to be united were cut 
across, and the exposed surfaces brought into contact and held 
in place by a coating of paraffin. In successful cases the parts 
united, at least so far as the integumentary organs are concerned, 
and the moths that subsequently emerged were made up of parts 
of two species. As a rule, each part developed its own specific 
coloring, showing no mutual influence of the parts on each other. 
In two cases, however, evidence of some influence was found. 
The posterior part of the abdomen of a female pupa of Callosa- 
mia promethea was united to the rest of the body of Samia 
cecropia. In the moth the promethea part assumed the color 
of the cecropia. In another union between Telea polyphemus 
and Samia cecropia the small piece of the latter assumed the 
color of the former. According to Mayer, the color in these 
moths is largely due to the drying of the hemalymph. The 
explanation of the two cases just given may be that some 
of the hemalymph of the larger component got into the smaller 
part. If the result is due to this, the outcome is only remotely 
connected with the “influence” of one part on another. 
In conclusion, it is apparent that grafted pieces of different 
species have no mutual influence on each other of the kind that 
characterizes species. In this respect the results are quite differ- 
ent from the effects of cross-fertilization; the difference is prob- 
ably due to the actual union within the same cell of the characters 
of the two parents in cross-fertilization, and to the absence of 
any such intimate fusion in the case of grafting. In other words 
the cells retain their specific characters in the unions by graft- 
ing, and although cells of different species may live side by 
side and form a part of a new individual, they have no influ- 
ence of a specific nature on each other. 
Special Problems of Development 
It has been pointed out that by means of grafting it has been 
found possible to study the question of self-development versus 
correlated development of parts of the embryo. In most cases 
