236 BOTANICAL GAZETTE [SEPTEMBER 
Wieland for Cycadoidea ingens. The fruit of the Bennettitales should be 
considered, not as a flower, but as an inflorescence. 
A diagram showing the relationships of great groups is submitted. From 
the Protopteridea, the ancestors of the Filicales, is derived a stock which 
becomes differentiated into two main lines, the Salisburiales and Cordaitales. 
At an early period the Cycadales were derived from the Salisburiales 
and, later, the Coniferales came from the same stock. From the Cordaitales 
at an early period came the Bennettitales and, later, the Gnetales and Angio- 
sperms. More must be known of the life history of fossil forms lying between 
pteridophytes and gymnosperms, and also of fossils in these two groups, 
before a satisfactory diagram of relationships can be constructed. CHARLES 
J. CHAMBERLAIN 
ACCORDING TO THE INVESTIGATIONS of J. Brzezinski, the canker 
disease of trees, long attributed to Vectria ditissima, is never caused by this 
fungus, which, in the opinion of the author, is merely a saprophyte on dead 
portions of the bark. Inoculations from pure cultures of Nectria failed to 
produce the disease. The author regards certain bacteria which he found 
growing in the wood as the true cause of the injury. Three species are 
described, viz., Bacterium mali, B. pyri, and B. coryli, growing respectively 
on apple, pear, and hazel. They differ but slightly in cultural characteristics. 
Inoculations of B. maiz and B. gyri in the wood of apple and pear, respec- 
tively, produced discolored areas which gradually extended for a period of 
several years, forming darkened lines in the wood. In three instances only 
were small cankers produced on appletrees. The author regards the canker 
wound merely as one of the external manifestations of the bacteriosis 
from which the tree is suffering. The disease may manifest itself also as 
general bacteriosis causing a sickly appearance of the whole tree and pro- 
ducing chlorosis in pear trees. Further, irregular knots on the limbs, a form 
of twig blight, and root knots are regarded as manifestations of the disease. 
The twig blight seems to resemble the disease produced on young apple 
twigs by Bacillus wis id ovorus Burri The root knots referred to are 
“crown galls.’”-— H. HASSELBRING 
PAUL discusses the functions of the rhizoids of mosses and announces Cer- 
tain conclusions, which the writer has held and taught for a number of 
years, as a result of his observation of the structure and development of these 
organs. Paul brings no experimental evidence, but relies on more extensive 
data of the same kind. He holds that the chief function of rhizoids is 
anchorage; as accessory functions he recognizes (1) the capillary storage and 
conduction of water by felted rhizoids, and (2) the absorption of water and 
“4 BRZEZINSKI, J., Le chancre des arbres, ses causes et ses symptomes. Bull. 
Acad. Sci. de Cracovie 1903: 95-143. pis. 2-8. 
5 PAUL, H., Beitrage zur Biologie der Laubmoosrhizoiden. Bot. Jahrb. System. 
32 5 231-274. 1903. 
