Apr. 24. 1916 Crowngall Studies 181 



made early; and they contain, along with the proliferating tumor cells 

 (blastomous cells), the same teratoid elements as the primary tumor. 

 These are true daughter tumors, being connected back to the primary 

 tumor by a tumor strand which is quite different both in structure and in 

 location (PL XXI) from that occurring in the Paris daisy. The latter, it 

 will be remembered, follows the line of the spiral vessels in the inner wood, 

 and is parenchymatous in its structure, containing only here and there 

 some vessels (scattered trachei) . This tobacco tumor strand occurs in the 

 cortex, consists almost entirely of vessels, and is a true stem (stele), 

 although developed under a pathological stimulus, and in a part of the 

 plant where no stele was ever seen before — namely, in the outer cortex, 

 through which it can be traced (parallel to the long axis of the stem) for 

 long distances and from which at intervals leafy tumors are sent to the 

 surface of the plant. I^rom its frequent proliferation in the form of 

 tumors it is evident that parenchymatous (blastomous) elements must 

 also occur in the strand, but they are not abundant. In fact, in the 

 parts I have examined they are almost as infrequent as are trachei in 

 the daisy strand. Cross sections and longitudinal sections of this re- 

 markable tumor strand show it to have spiral vessels in its center, sur- 

 rounded by trachei cut by ray cells, beyond which is a cylinder of 

 cambium surrounded by a cylinder of phloem, containing well-developed 

 sieve tubes. This tiny stele has no cortex or epidermis because it does 

 not need any, being surrounded and sufficiently protected by the normal 

 cortex of the tobacco stem. This is a phenomenon due apparently to 

 my new manner of inoculation (into shoot anlage), because some years 

 ago by inoculating intemodally on tobacco stems I obtained and figured* 

 tumors and a tumor strand in cortex corresponding to those found in the 

 Paris daisy — that is, composed chiefly of small-celled parenchjona. 

 The difference in results must therefore be due to difference in the 

 kind of tissue inoculated, each developing pathologically according to 

 its own growth tendencies. 



Fourth, on some plants (which were tobaccos) I have also obtained 

 leafy tumors by making my bacterial inoculations in places where no 

 bud anlage are known to exist — for example, in the middle of leaves. 

 Ordinarily when leaf tissue in tobacco grows, it only produces more leaf 

 tissue;^ but when the crowngall organism (hop strain) is pricked into 

 midribs or side veins, tumors arise and a portion of them are leafy — that 

 is, bear shoots. I have obtained 27 such leafy tumors on a single plant 

 and several on a single leaf, all within a period of a few weeks (PI. XXIII). 

 It is easy to obtain them. The young leaves yield a larger proportion 

 of such tumors than the older ones, and I have observed no shoot-bearing 

 tumors on leaves which were fairly well developed when inoculated. 



• Smith, Erwin F., Brown, Nellie A., and McCulloch, Lucia. The structure and development of crown- 

 gall: a plant cancer. U. S. Dept. Agr. Bur. Plant Indus. Bui. 253, pl. 102-103. 1912. 

 2 I have never got any leaf cuttings of it to take root. 



