1»17] DISEASES OF PLANTS. 557 



amount of available water, degree of exposure to wind, and possibly some factor 

 affecting nutrition. 



On the dieback disease of Paulownia tomentosa caused by a new species of 

 Valsa, T. Hemmi (Bot. Mag. [Tokyo], SO {1916), No. 557, pp. 80^-315, figs. 4; 

 abs. in Mycologia, 9 (1917), No. 1, p. 40). — Since 1914 tlie autlior has studied a 

 dieback of P. tomentosa, due to a fungus described as a new species, V. 

 j'auloionice, and said to be an almost omnivorous saprophyte. It attaclis all 

 parts of these trees of any age, but only at points where they have been injured, 

 producing a layer of dead cells. Winter injury constitutes an important con- 

 tributing factor in the vicinity of Sapporo, Japan. 



Diseases of woody plants in North Africa, R. Maire {Bui. Sta. Forest. Nord 

 Afrigue, 1 {1916), No. 4, pp. 121-130, pi. 1, figs. 4). — Arbutus unedo in Algerian 

 forests is subject to a disease producing a crowding together of the leaves, 

 which persist after they become dead and dry, giving the appearance of a 

 small witches' broom. The trouble is said to be due to a fungus described as 

 Exobasidium unedonis n. sp., which is further discussed. The same branches 

 as bear the above fungus show another fungus described as OloBOsporium con- 

 viva n. sp., and a third designated as Phoma arbuti n. sp. Rosa seinpervirens 

 near Algiers is frequently attacked by a fimgus thought to be new, and 

 described as Phragmidium roscB-semperxnrentis. 



The influence of certain climatic factors on the development of Endothia 

 parasitica, N. E. Ste\'ens {Amer. Jour. Bot., 4 {1917), No. 1, pp. 1-32, figs. 3).— 

 The author made a study during 1914 and 1915 of the behavior of chestnut 

 blight on trees inoculated with E. parasitica. The experimental tracts ranged 

 from Concord, N. H., to Charlottesville, Va., this stretch of country including 

 a transition region for several important plant diseases which are named. 



The average annual growth laterally of the chestnut cankers was found to 

 increase at 11 stations trending from the north to south, but to decrease with 

 elevation at stations in the Catskills. Rainfall seems to bear no relation to 

 canker development. The fvmgus can resume growth at once on return of 

 atmospheric warmth. No relation of climatic factors to pycnospore develop- 

 ment could be traced. Ascosporea were not developed in some cases ^vithin 

 18 months after inoculation. No mature perithecia developed during 1914, 

 but both perithecia and ascospores appeared abundantly at many stations in 

 the (late) winter, spring, and summer of 1915, atmospheric moisture appearing 

 to have considerable connection, temperature much less, with ascospore devel- 

 opment. Perithecia were first observed in the spring during a period of high 

 humidity in 1915. Dry weather apparently tends to reduce the spread of the 

 chestnut blight by lessening spore production. 



The results of the present investigation are said to agree with those reported 

 by Rankin for Ulster County, N. Y. (E. S. R., 32, p. 54). The data here 

 presented are thought to indicate a probability that chestnut canker wUl 

 spread more rapidly in the Southern than in the Northern States. 



The influence of temperature on the growth of Endothia parasitica, N. E. 

 Stevens {Amer. Jour. Bot., 4 {1917), No. 2, pp. 112-118, fig. J).— The observa- 

 tions above noted have been continued, though abandoned at all but six 

 stations on account of the increased prevalence of the disease, and the results 

 are here tabulated and discussed. 



The lateral growth of the cankers of E. parasitica on Castanea dentata, as 

 noted at these stations for the year ending with May, 1916, as for the previous 

 year, was apparently not influenced by the amount or frequency of rainfall. 

 The rate of growth of the fungus agi'ees closely with the temperature for the 

 period, as computed by the system of remainder summation indices, or that of 

 exponential summation indices, and less closely but stiU in a general way with 



