24 FARLOW ON THE GYMNOSPORANGIA 



the case, our common G. dampen must be regarded as a variety of Q. Juniperi, but, as tbe 

 matter now stands, I must believe that the two are distinct, and that the existence of G. 

 conkum in the United States rests only on a few specimens resembling G. clavip)es in habit, 

 but which, as far as can be made out from spechnens which as a whole are in poor condi- 

 tion, have longer and slenderer spores on pedicels which are not perceptibly thickened 

 below the spores, and whose promycelia are in twos or fours near the septum. 



RoESTELiA Rebent. 



Aecidia usually hypophyllous, lower part sunk in the swollen tissues of the leaves, form- 

 ing, above, cylindrical, conical, or oblong projections which are often split and fringed in 

 the upper part, peridium composed of large, colorless cells, spores brownish or orange- 

 colored, subglobose when mature, formed in moniliform rows. Spermogonia punctiform, 

 forming minute dark-colored pustules in discolored spots on the upper surface of the 

 leaves. Mycelium infesting the leaves and stems of diiFerent Pomeae. 



The old genera Aecidium, Roestelia, and Peridermium canuot be distinguished from one 

 another except in an arbitrary way. The species of Peridermium are parasitic on differ- 

 ent Coniferae, the Roesteliae on species of Pomeae, and Aecidium proper is very widely 

 diffused. Wolff" ^ considers that Peridermium Pini is the aecidial form of Coleospormm 

 Senecionis, and De Bury and Hartig have connected other Peridermia with Chrysomyxa 

 and Calyptospora. The Roesteliae differ from the species of Aecidium in the fact that 

 the peridium is elongated in a more or less tubular form, whereas in Aecidium it is 

 short. But in forms like B. p)^nicillata (Sow.) the peridium is comparatively short, while 

 in Aecidium Fraxini Schw. the peridium is so long that in the Syn. Fung. Am. Bor. it 

 was placed by Schweinitz in Roestelia. In his work, Untersuchungen iiber die Brand- 

 pilze, De Bary considered it to be a distinguishing mark of Roestelia that the spores were 

 not formed from all the cells of the sporiferous filaments but from every other cell, so that 

 the spores hung together for a short time by the shrivelled sterile ceUs. Reess adopts the 

 same view, but more recently De Bary^ has stated that similar sterile cells are found in 

 other genera than Roestelia and they are certainly found in Caeoma luminata Schw. 

 and in species of Aecidium which I have examined. The cells which form the peridium 

 are, like those found in Aecidium, large and colorless, with thick walls which 

 generally have peculiar markings. They are only loosely adherent, and although they 

 may cohere to one another in longitudinal rows, the rows, especially at the upper end of 

 the peridium, soon separate from one another and form a fringed mouth to the perid- 

 ium. In some species, however, the cells at the apex remain united and those below sep- 

 arate from one another so as to form a sort of lattice-work, through the meshes of which 

 the spores escape. 



The spores of Roestelia are more or less angular, when young, from mutual pressure, 

 but when mature they generally become globose. They are almost always of a brownish 

 color, but hi one of our species they are orange-colored. The wall of the spores is double, 

 consisting of a rather thick endospore and a thin exospore. The cndospore is perforated 



' Beitrag zur Kentniss der Scbmarotzerpilze, Landwirtlisch. ' Bot. Zeit., 1SC9, p. 786. 



.Jabvb., 1877. 



