710 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



tthutt., on Barberry nvith Puccinia Arrhcnathcri, Erikss., on Tall Oat-grass. 



XVI. Experiments with Puccinice on Phalaris ; reports new observations 

 on the connection between these and jecidia on Liliacea and Orchidece. 



XVII. Oymnosporangiv/m experiments ; notes on infection of Rosacea. 



W. G. S. 



Rusts. 



Rusts of Cereal Crops: Origin and Propagation. By J. 



Eriksson (Ann. Sr. Nat. Sot. xiv. 1901, pp. 1-124 ; xv. 1902, pp. 1-160 ; 

 figs. 10, plates 5). — For twelve years the author has carried on an 

 Important research on the rust fungi which damage Corn crops. Many 

 papers and a book have been issued on his work. Reviews of the earlier 

 work have appeared in English periodicals, e.g. Plowright (Gard. Chron., 

 1898), Massee (Natural Science, 1898), Smith (Nature, August 1898). 

 The chief object of the present paper is to furnish evidence in favour of 

 the author's view that rusts of cereals are propagated by a germ which 

 inhabits the grain. The paper is diffuse, and only a brief outline can be 

 attempted. 



Part I. — Observations and investigations which suggest an internal 

 origin of rust. A. Observations in the open field (xiv. pp. 1-124). The 

 following are noteworthy. On autumn-sown Wheat, uredo patches were 

 never seen by Eriksson earlier than thirty to thirty-four days after sowing 

 the grain ; yet uredospores under observation cause infection in ten to 

 twelve days. Although the teleutospores germinate in April or May, the 

 first appearance of uredo patches in the years 1891 to 1897 was : 

 (a) autumn-sown Rye, last week of June to first and second weeks of 

 July; (b) autumn-sown Wheat and Barley, second or third week of 

 July; (c) spring Wheat, Oats, and Barley, second week of July to 

 beginning of August. Again, autumn Wheat and Rye, 100 metres 

 distant from the nearest Barberry bushes, showed uredo patches before 

 spring Wheat and Barley only fifty metres off. The same variety of 

 Wheat was sown at intervals of fourteen days from April to June ; in each 

 case the first trace of rust appeared in eighty-one to ninety-six days. 

 Varieties of Oats, Barley, and Wheat are unequally disposed to rust. 

 These and other results are given, and Eriksson says they can only be 

 explained by assuming an internal germ as the cause. The remainder of 

 Part I. deals with the different species and forms of rust recognised by 

 Eriksson, and with the course of their development. 



1. B. (xv. pp. 1 50.) Experiments with isolated cultures. Cereals 

 liable to rust were grown in closed chambers and protected from outside 

 infection. At first the results were negative, because healthy plants could 

 not be obtained in the chambers, but as technical details were improved 

 normal plants were obtained. The majority of the experiments showed 

 that plants protected from external infection can develop rust, of which 

 the only source is an internal germ in the seed. The different methods of 

 protecting the plants are shown in figs. 1 8, and plates 1 and 2. 



Part [I. (pp. 51 129). Morphology and biology of the internal germ 

 of rust. Thifl is the crux of Eriksson's work, since it is the source to 

 which he ascribes the propagation of rusts in epidemic form year after 

 year, lie criticises the opinion that the rust may be propagated by 



