August 9, 1900] 



NA TURE 



353 



teleutospores of the Puccinia stage hibernate and in the follow- 

 ing spring germinate, producing secondary spores (also known 

 ,is sporidia), which infect barberry foliage and give rise there to 

 the Aecidium stage with its aecidiospores ; aecidiospores do not 

 infect barberry again, but on Gramineae produce the uredospore 



i* !•"•• I. — IMack rust of oats ( Puccinia graminis, spec, form ATrnac). a, leaf, 

 and /', cluster of fruits of barberry with Aecidium herberidis (nat. size) 

 c, leaf, and d, a spikelet of oat with Uredo stage (nat. size) ; e, uredo 

 spores ( X 500) ; y", sheath of oat with Puccinia stage (nat. size) 

 g, teleutospores ( X 500). (J. Eriksson.) 



stage, thus completing the cycle. Accompanying the aecidium- 

 cups there occurs constantly a form of reproduction, the spermo- 

 gonia, which gives off spermatia or spore-bodies whose function 

 neither De Bary, nor any one since, has been able to determine. 



These results had an important and direct bearing on rust- 

 epidemics of cereals, and they gave an impetus to further 

 research on the biology of the whole group of rust-fungi or 

 Uredineae, and, in fact, of all other parasitic fungi. In 1889, 

 twenty-five years after De Bary's first results, both Plowright 

 {loc. cit., p. 56) and Rostrup published a list of fifty heterrecious 

 rust-fungi. Recently, Dietel (Engler's " Pflanzenfamilien ") 

 gave about a hundred cases, including species outside the 

 Uredineae. Works like Plowright's "British Uredineoe" are 

 the evidence of this impulse, and a perusal of current botanical 

 periodicals shows that the subject is by no means exhausted. 



Ten years ago, three species of rusts occurring on crops of 

 cereals were recognised : 



(i) Puccinia graminis, with its Aecidium stage on barberry 

 and mahonia ; its uredospore and teleutospore stages on wheat, 

 barley, oats, rye, and about a hundred species of grasses (see 

 Fig- I). 



(2) Puccinia rubigo-vera, with Aecidium on many species of 

 Boragineae ; uredospores and teleutospores on wheat, rye, and 

 a number of grasses (see Fig. 2). A variety, j/w/Z^jr, was dis- 

 tinguished on barley. 



(3) Puccinia coronata, with Aecidium on species of buckthorn 

 {Rhamnus) ; uredospores and teleutospores on oats and several 

 grasses. 



The four important European cereal-crops were thus known 

 to have each two forms of rust-fungi, distinguished in their 

 external characters, and with distinct Aecidium host-plants. 

 Yet it was by no means certain that epidemics of rust were 

 fully traced out. Fortunately the economic importance of 

 rust-epidemics was enough to enforce attention from State 

 departments, notably in Sweden, United States of America, 

 Australasia, and in various parts of Europe and other coun- 

 tries. In Biitain, while good work has been and is still done on 

 rust-fungi, there has, in recent times, been no specially organised 

 research relating to cereal rusts, probably because recent 

 developments on the subject have rendered a research too exten- 

 sive for the resnurces of any but workers specially retained and 

 remunerated. The investigations on rusts of cereals reviewed 

 here are mainly the outcome of State-aided research. 



In Sweden, the Government in 1890 offered ten thousand kroner 

 (about 560/.) to the Royal Swedish Academy of Agriculture 

 for an investigation on rust of wheat, &c., intended, at first, to 

 extend over three years, but which has been continued up till 

 now. The grant was placed under the control of Jakob 

 Eriksson, now professor of vegetable physiology at the experi- 

 mental station of Albano, near Stockholm. The experiments 

 were started in 1890 ; the first important results of Eriksson 

 and his co-worker, E. Henning, appeared in 1894 {Zeitsch. f. 

 Fflanzenkrankheilen, iv. 1894, pp. 66, &c.), and as a bulky 

 volume in 1896 ("Die Getreiderost," Stockholm, 1896; 463 

 pp. and 14 plates). Other contributions, and re-statements of 

 lormer work, have been made by Eriksson in almost every 

 existing botanical periodical. The present summary is based 

 chiefly on the latest re-statement {Kevue ght. d. Sciences, ii. 

 January 15, I9CX3, pp. 30-39), with aid from other papers.' 



In Eriksson's experiments test-plants of cereals were grown 

 from seed, or young plants from the open were transferred 

 into pots. The soil used was generally sterilised. After 

 inoculation with rust the plants were watered with distilled 

 water, placed under large glass bell-jars moistened with distilled 

 water, and left undisturbed for twenty-four hours in glass-houses 

 specially constructed for the experiments. After this, observa- 

 tions were made at frequent intervals. The main lines of 

 investigation were : (i) to define the species which cause rusts 

 of cereals and grasses, and to trace their life-history ; (2) the 

 propagation of the rusts ; (3) germination and vitality of the 

 various forms of spores. 



According to Eriksson's results, the three species and one 

 variety of rusts attacking cereals and grasses as recognised in 

 1890 really represent twelve species and many subdivisions. 

 His list is as follows, but the less important host-plants amongst 

 the grasses are omitted : — 



Species l. Puccinia graminis. Pers. (Black Rust), with 

 Aecidium herberidis. Specialised form ( i ) Secalis, on Secale cereale 

 (Rye), Hordeum vulgare {Ba.x\ty), H. jubdtum, Triticum repens, 

 T. caninum, &c., Elymus arenanus, and Btomits secalinus. 

 (2) Avenae, on Avena saliva (Oat), A. elatior, &c., Dactylis 



1 J. Eriksson : Ber. d. deutsch. boian. Gcss., 1B94, p. apa ; 1897, p. 183. 

 Jahrbuch f. 7viss. Botanik, xxix. 1896. Botan. Centralflatt, Ixxii. 1897, 

 pp. 321-5 and 354-62. Centralblatt f. Bakter. u. Parasitenkunde, Abt. ii. 



1897, pp. 291-308. 



NO. 1606, VOL. 62] 



