12 FARLOW ON THE GYMNOSPORANGIA 



The spores of G. EU'isii are very striking and differ from those of the rest of the genus 

 in being very long and narrow and in being usually more than two-celled, the most usual 

 numbers being three and four. The amount of jelly in the sporiferous masses is less than 

 in other species, and in consequence dried specimens give a better idea of the fungus as it 

 appears in nature than is generally the case in the present genus. The promycelia are 

 very abundant and very short, the lower sterile part found in other species being almost 

 wanting and the part bearing the sporidia being much curved, so that the promycelia com- 

 ing from the cells of one spore sometimes wind round and enclose another spore, making 

 dissection difl&cult without tearing off the promycelia. One not unfrequently finds spores 

 in which the upper cell is more or less deeply cleft, as in PI. 2, fig. 17. 



In spite of the fact that in certain details, G. Ellisli differs from the majority of the 

 other species of Gymnosporangium it seems to me that Kcirnicke ^ is not warranted in 

 establishing a new genus Hamaspora, founded on two species, G. Ellisii growing on 

 Cupressus thyoides and Phr'agmidium longissimum Thlim. growing on Rubus rig'idus 

 at the Cape of Good Hope. In the first place, the gelatinous substance is not wanting in 

 G. Ellisii, as can easily be seen in examining fresh specimens, and furthermore, 

 the fact that the spores are more than two-celled is equally true of G. hisejitatum, a 

 species which imdoubtedly belongs to Gymnosporangium. On the other hand, in H. long- 

 issima Kornke., admitting that the teleutospores bear a great resemblance to those of 

 G. Ellisii, the specimen in Mycotheca Universalis, No. 542, shows an abundance of uredo- 

 spores surrounded by the circle of large paraphyses generally found in the uredo-spots of 

 Phragmidium, while in G. Ellisii there are no uredo-spores at all. When we consider 

 also that the species of Phragmidium generally occur on species of Rubus or related gen- 

 era, and Gymnosporangium only on Coniferae, it would certainly seem that M. longissivia 

 should be kept in Phragmidium where it was placed by Von Thiimen, and that G. Ellisii 

 should be retained in Gymnosporangium. I am perfectly willing to admit that the last 

 named genus approaches the former, but the matter is not helped by creating a third 

 genus less clearly marked than either of the others. 



The present is more limited in its range than our other species, as far as at present 

 known. It probably has often been overlooked, on account of its small size, and may 

 occur wherever the white cedar, Ciq^ressus thyoides, is found. It is certainly common in 

 such localities in Massachusetts, and in passing from Boston to Washington by railroad, 

 I have seen the peculiar distortions along the whole route wherever the white cedar 

 occurred. 



Gtmnosporangium clavaeiaefokme De Cand. 



Gymnosporangium clavariaeforme D. C, Flore fran^aise, Vol. ii, p. 217 ; Reess, loc. cit., p. 



21. Exsicc. Ellis, North American Fungi., Fasc. iii, 273. 

 Podisoma clavariaeforme Duby, Bot. Gall., Vol. ii, p. 881. Oersted, Nouveaux essais de 



semis. PI. 3 and 4. 

 Podisoma Juniperi communis Fr., Syst. Myc, Vol. iii, p. 548. 



1 Hedwigia, Vol. xvi, p. 22. 



