NEOPLASMS 469 
ful study failed to reveal any parasitism as the cause of 
the growths and since the first two were of slightly vary- 
ing structure it is not believed that one is a metastasis 
from the other. 
The sex incidence stands in direct relation to the pro- 
portion of total males and females posted or in other 
words it is the same for the two. The figures might be 
somewhat affected were the gender of all the parrakeets 
available but the tumors growing in the upper renal area 
frequently destroy the sex gland. 
Definite statements concerning the importance of 
breeding in the causation of neoplasms cannot be made 
since we cannot quote fignires for the percentages of wild- 
and captive-born of our entire autopsy list. The data 
are confused by scanty information concerning the 
twenty-six parrots, the history of which is vague and I 
am perhaps too severe in accrediting the birth of sixteen 
of them to captivity. This was done because of a lack of 
exact information concerning these specimens and, be- 
cause their variety is kno^vn to breed when captive by the 
residents of their habitat (4), the distribution into wild- 
and captive-born is based upon what information we have. 
If the order Psittaci be subtracted entirely, it leaves a 
total of 62 tumor-bearing animals of kno^vn breeding, 49 
of which were born in the wild, thirteen in captivity, a 
fact which strengthens the thought that unnatural breed- 
ing increases the chance of neoplasms. 
The known length of capti\dty has also a direct bearing 
on this point. The figures' given in the columns " known 
captivity " and " average for tumor bearers " were 
compared with figures obtained by averaging the lives of 
fifty others (when possible) of the same order or of at 
least three times as many as bore tumors. Animals 
dying from injury were excluded. With one exception 
the average for " tumor bearers " exceeded that for 
*' non-tumor bearers"; the exception, the Ungulata, had 
(4) See Gould's Birds, Vol. II, p. 83. 
