850 REPORT— 1903. 



nature of a vacuole and to be intimately concerned with the nutrition of the 

 nucleus, which at this time is of a remarkably large size. The number and size 

 of asci in the ascocarps are very variable. Under favourable conditions as many" 

 as fifty have been counted, while in other instances only one is present. Nutrition 

 seems to be the principal factor in this variation. The walls of the asci are 

 usually thin, but desiccation causes a thickening, in some cases as much as 5 fi. 

 The dehiscence takes place by a small cap or lid at the apex of the ascus, 

 the spores being collected into a mass in this region and expelled explosively. 

 The spores vary in number and in size in different asci. Normally more than 

 two hundred are found, but as few as sixteen have been seen. 



These results point to a close relationship between the genera Ryparobius and 

 Thelebolus, the structure of the archicarp in the species here described being 

 practically identical with that figured by Brefeld for the latter genus. Hence the 

 position of Thelebolus among the Hemiasci, asserted by this author, is incorrect, 

 and it must be regarded as a true Ascomycete. 



6. Culture Experiments ivith Biologic Forms of the Erysiphaceae. 



£y E. S. Salmon. 



Within the last two years the existence of ' biologic forms ' in the Eri/siphacem 

 has been definitely proved. The special restriction of infection-power is found not 

 only in the conidial {Oidiian) stage, but also in the ascigerous stage. 



In the present paper the author gives the results of experiments, which show 

 that the infection-powers of the conidia which obtain when uninjured growing 

 leaves of a plant are inoculated become altered under certain cultural conditions 

 when cut-ofl' leaves are used. The experiments prove that under the above con- 

 ditions («) a ' biologic form ' which in nature is restricted to the species of a 

 certain genus of host-plants becomes capable of infecting species belonging to 

 another genus, and {b) species of plants which are immune in nature are able to 

 be infected. 



Among a number of detailed experiments given the following may be mentioned 

 to illustrate the above points : (a) The conidia of the biologic form of Erysiphe 

 Graminis on wheat, which has been repeatedly proved to be unable to infect 

 barley when sown on uninjured growing leaves of the latter, proved capable of 

 doing so when sown on cut-oft' leaves under certain cultural conditions. (6) The 

 two species of wheat, Triticum dicoccum and T. monococcum, are, under natural 

 conditions, immune against the attacks of E. Grcnirinis. Under the above- 

 mentioned method of culture, however, the conidia of E. Graminis on wheat are able 

 to infect both species, producing conidiophores and ripe conidia in six to seven 

 days. 



The author considers that it is possible that iti this change of infect ion-potoers 

 of biologic forms of 2)arasitic fungi in consequence of injurg to the host-plant an 

 explanation may be found of the sudden appearance of disease on plants hitherto 

 immune. Attacks by animals, or injury by rain, frost, &c., may produce the same 

 change in the leaf-ceils or the host-plant as that brought about by the above 

 method of culture, and consequently render susceptible to the attacks of a fungus 

 plants otherwise immune. 



7. Willow-canker. By Professor T. Johnson, D.Sc, F.L.S. 



Considerable damage has been caused by a canker in an osier holt in the west 

 of Ireland. The bark looks burnt and blistered at the canker-spot, shows black 

 specks breaking through it, and gradually peels oft' leaving the wood exposed 

 and sometimes shredded. The microscope shows abundant mycelium present in 

 the pith as well as the rest of the shoot. The black specks are the perithecia 

 of Physalospora {Botryosphceria) gregaria, Sacc, or in some cases the pycnidia of 

 J)endrophoma Therryana, Sacc, which Saccardo thinks may be the spermogonium 



