138 The American Naturalist. [February, 



magnificent infection experiments, but now for the first time we have 

 full details of all the laboratory and field investigations. In the limits 

 of this review it will be possible to notice only the first of these two 

 volumes. This forms Heft XI of the Untersuchungen and is entitled 

 Die Brandpilse II. It deals principally with infection experiments 

 and gives in full the results obtained with Udilago Carbo on oats, U. 

 omenta on sorghum, and U. maydis on maize. These experiments were 

 carried on through a period of four years with striking results and in 

 case of corn, with most unexpected ones. Space forbids entering into 

 much detail. Those who wish for details will naturally consult the 

 volume itself. Suffice to say that the infective material consisted of 

 the yeast-like conidia propagated in nutrient solutions made from fresh 

 horse dung. 



In case of oats the best results from direct infection were 17 to 

 20 per cent, of smutty plants, obtained by spraying during the 

 earliest stage of germination. Infections made when the embryo was 

 one cm. long gave only 7 to 10 per cent of smutty plants; when it was 

 2 cm. long (500 plants), only 2 per cent became smutty. When the 

 plumule had pushed through the enfolding sheath scarcely any of the 

 plants could be infected, 200 seedlings in this stage yielding only 1 per 

 cent of smutty plants and 200 more remaining entirely free. The in- 

 fections took place through the young axis and also through the sheath- 

 ing leaf so that both Wolff and Kuhn were right, but a majority of the 

 infections were through the young axis. In a second series of experi- 

 ments garden earth was sprayed with the smut conidia and two days 

 later oats were planted 1 cm. deep and subsequently transplanted to the 

 open field : 300 of these seedlings yielded 5 per cent of smutty plants, 

 and 300 more, 4 per cent., i. e. a much smaller per cent than was anti- 

 cipated. In a third series fresh horse dung was mixed with garden 

 earth which was then abundantly impregnated with the smut conidia. 

 Three days later oats which had been soaked but were not yet germin- 

 ated were planted in this soil at a depth of scarcely 1 cm. These seed- 

 lings were divided into two lots, 300 were kept for a time in the labor- 

 atory at a temperature of over 15°C, and 300 were placed in the cellar 

 where the temperature did not exceed 7°C. Of the 300 kept in the 

 laboratory 27 to 30 per cent finally became smutty ; of those kept in 

 the cellar, where germination proceeded more slowly, 40 to 46 percent 

 became smutty. This shows clearly that fresh horse dung greatly 

 favors the development of smut and that weather which retards germi- 

 nation is also favorable. In the fourth series of experiments the infec- 

 ftted for a loner time arti- 



