Jan. 29,1917 
Mechanism of Tumor Growth in Crowngall 
183 
Similar results were obtained on cauliflower leaves by exposure to 
the vapor of secondary methylamine (PI. 46, 62, 63, 64). 
Also on cauliflower plants exposed to the vapor of primary ethyla- 
mine the younger leaves showed distortions and thickenings of the 
parenchyma with changes in the. palisade tissue quite suggestive of 
curly top of sugar beets, mosaic of tobacco, and similar diseases of 
undetermined origin but suspected of being due to parasites (Pis. 63, 
64). The etiology of these obscure diseases can perhaps be unraveled 
by exposure of healthy plants to the products of a variety of micro¬ 
organisms and here, possibly, is a clew. 
Consequently without entering into further details I will say that I 
think I have established my original hypothesis—viz, that dilute ammonia 
causes intumescences and have rendered it probable that ammonia 
liberated within the cell in small quantities by the imprisoned bacteria 
must be one of the causes of the excessive and abnormal cell proliferation 
in crowngall. Probably amin compounds also help to determine it. 
Since an acid and an alcohol are likewise produced by the crowngall 
bacteria and this alcohol and this acid (as well as many other acids) in 
pure dilution and also in combination with ammonia caused galls or 
intumescences in my experiments, the acid (or acids), the alkalies,, and 
the alcohol must, I believe, act together in producing the tumors, and 
osmotically rather than chemically. 
To return now to the experiments as a whole, here is abundant further 
evidence that changes in stimulus can produce changes in structure, if any 
such additional evidence were needed, considering the multiplicity of 
arguments from regeneration experiments, especially on animals. And 
this time the changes, some of which if more regular (and there is a reason 
for that) are nevertheless as striking as anything observed in beginning 
neoplasms, have been produced, in the absence of bacteria, with known 
chemical substances—some of them the by-products of the metabolism 
of a cancer parasite, provided, of course, that crowngall is admitted into 
the family of the neoplasms, as I think it must be (18, 19, 20). 
out of ordinary cells being in some way brought about by a closely attached but external parasite which 
feeds upon them as they grow by thrusting its mouth parts into them. These cells are often 20-30 nucleate 
and a hundred times as large as the normal cells, so that often in sections of the galls they have been mistaken 
for parts of the parasite, being quite unlike anything ever observed in thenormal tissues of thehost. Here 
the stimulating substance, which might well be some ammonia compound from the urine or feces of the 
feeding larvae, not only stretches the cell wall, as in 1 and 2, but that or some other substance also causes a 
fission of the nucleus. But fission of the nucleus may occur in crowngall without hny marked enlargement 
of the cell, and also cells in that tumor may show hypertrophy without developing multiple nuclei. When 
discussing my results with Dr. N. A. Cobb, of the Bureau of Plant Industry, he called my attention to 
certain facts respecting the orange-root nematode of Florida, which seem to me to be very pertinent in this 
connection. This eelworm, which is the type of his new genus Tylenchulus (16,17), does not produce galls, 
although it is closely related systematically to the gall-forming eel worms of the genus Heterodera. This 
striking difference is correlated with the fact that the larvae of Tylenchulus are hatched free in the earth 
and never insert into the roots of -the orange plant anything but their head parts, all their anal excretions 
being voided into the earth, whereas the larvae of Heterodera are entirely buried in the plant from the 
beginning, so that all their excretions must necessarily come into intimate contact with the growing tissues 
of the host. It would seem, therefore, that nematode galls must be due to anal rather than buccal or skin 
excretions. 
