OF ARTS AND SCIENCES. 267 



the case with the RoesteUa obtained from macropus, where the peni- 

 cillate habit was preceded by one very similar to that obtained in the 

 present instance, the RoesteUa producing slender unbroken peridia, 

 which became lacerate and curled backwards only after exposure for 

 a short time to rain and wind. The same absence of any laceration 

 was marked in the case of the aurantiaca culture, where the usually 

 lacerate peridia were unbroken. 



Of these three forms, the first, ic, seems to me to be the true lacerata, 

 as understood by Oersted and distributed as above mentioned. As re- 

 gards the other two, no definite statement can be made. It should be 

 noted, however, that, with the exception of a slight variation in the 

 size of the spores, neither of these forms can be satisfactorily separated 

 microscopically from R. cornuia, as it occurs in this country on Fpus 

 Americana and Amelanchier. or from a form with the cornuta habit, col- 

 lected at Kittery, Maine, on Pyrus arhutifolia^ apparently a new host 

 for this species. Moreover, the form y often assumes a habit identical 

 with the most typical cornuta, if somewhat more slender. This I have 

 frequently observed at Kittery, and it is well shown by specimens col- 

 lected in Massachusetts by Mr. Seymore, who has kindly allowed me 

 to examine the RcestelicB in his herbarium. The possibility that these 

 forms y and z are one or both cornuta, is a natural inference from 

 the above facts. Yet, as already remarked, it is impossible from our 

 present knowledge of them to consider it more than a conjecture. 



Turning for a moment to my culture of G. conicum, it should be 

 observed that the RoesteUa obtained had the typical cornuta habit, 

 yet was not separable microscopically from the form y, while the date 

 of development corresponds with neither, being earlier by about two 

 months or more. This rapid development is therefore not easily ex- 

 plained, as the RoesteUa is not referable to any other form. That it 

 was accidental is rendered more probable from the fact that I was 

 unable to find any such RoesteUa during June in localities where 

 G. conicum was abundant. 



No microscopic examination of the spermogonia obtained was at- 

 tempted, and in general they presented much the same appearance. 

 Those from clavijjes, however, were the least conspicuous and those 

 from macropus were tinged with greenish and preceded in every case 

 by yellowish discolorations. G. globosum produced the brightest and 

 most luxuriant spermogonia of all, yet, as already stated, though ap- 

 pearing on four distinct hosts, they produced no gecidia, and it is impos- 

 sible even to guess with what RoesteUa it is connected. 



As regards the different species of Gymnosporangium, little need be 



