284 



THAXTER. MONOGRAPH OF THE LABOTJLBENIACEjE. 



fertile branch is unable, or barely able, to reach its base, a curious variety of conditions may result; the 

 most striking difference being indicated by such a form as is illustrated in fig. 8, Plate XLI, in which the 

 whole habit of growth of the secondary receptacle is changed, becoming identical with that of types like 

 H. chwtophilits (fig. 14). A form which has barely reached the base of the bristle is represented in fig. 

 10, and in this instance a rudimentary shield has been formed, while in fig. 9, one more fully, though 

 somewhat abnormally developed, is present. Enough transitional conditions have been examined to make 

 it evident that figs. 7 and 8 represent individuals of the same species. The existence of such remarkable 

 variations in a single species indicates that it is necessary to use considerable caution in determining the 

 limitations of species in this genus. 



The male individuals vary greatly in their development; some, like H. chatophilus, producing only 

 one or two antheridia directly from the typical four-celled primary receptacle; while others may show 

 much greater complication; the most highly developed type being illustrated by II. Ecfobiw, the second- 

 ary male receptacles of which may produce several hundred antheridia. The latter, in all species are 

 similar to others of the simple type, usually elongate and often showing no very marked distinction between 

 the neck and venter. In a single specimen of II. Paranensis the development of antheridia from a 

 single receptacle has been observed; an abnormal condition which finds a parallel among hermaphro- 

 dite forms, in which antheridia occasionally replace perithecia, as has been already illustrated in my 

 former Monograph. 



The generic type of Herpomyces appears to be very well defined, and although in II. Ectobice the 

 presence of creeping secondary receptacles in the male, as well as in the female, tends to set this form 

 apart from the others, it should be remarked that the same condition occurs in H. Anaplectce. Such male 

 receptacles in this species are, however, rudimentary, and the antheridia are generally produced from the 

 primary receptacle. The immediate affinities of the genus are not apparent, and there appears to be no 

 obvious connection between it and the other dioecious genera with simple antheridia (Amorphomyces and 

 Dioicomyces). Although the habit of producing fertile branches which grow horizontally from the primary 

 receptacle is found in other dioecious genera like Dimorphomyces, there is no other instance in the whole 

 group in which nutrition is effected by means of secondary attachments to the host. It is not difficult, 

 however, to imagine how a form like Dimorphomyces Thleopora, for example, might originate a somewhat 

 similar condition through the production from its fertile branch of rhizoidal attachments; since it already 

 lies in contact with the surface of the host. Among the known dioecious forms, however, which are char- 

 acterized by simple antheridia and with which the members of the present genus would be naturally supposed 

 to be more closely associated than with any of the other genera, one finds no tendency toward the formation 

 of such fertile branches, and their other characters vary widely from those of Herpomyces. The struc- 

 ture of the perithecium and the relatively greater complication in the general structure of both sexes might 

 be assumed to place it higher in the scale than either Amorphomyces or Dioicomyrrs, although the occur- 

 rence of a series of forms on Blattidse, which are supposed to be representatives of one of the most ancient 

 types of true insects, might perhaps have been expected to be correlated with a more primitive type in the 

 parasite. But although the unisexual forms with simple antheridia might for some reasons be assumed 

 to be the more primitive, the present genus is distinguished by a far more complicated structure than the 

 other unisexual forms of the same type, and Amorphomyces still remains, as has been pointed out above, 

 by far the simplest in structure of all the Laboulbeniales. 



Although I have as yet been unable definitely to determine this point, the number of spores in an 

 ascus appears to be eight, and if this is the case, it is a character shared only by Moschomyces and Comp- 

 somyces. It may be of interest to note that these genera present a further resemblance in possessing more 

 than usually numerous wall-cells in their perithecia. In one other respect the present genus is unique, 

 namely in the possession of a peculiar blackish foot-like process near or beside the terminal spinous proc- 

 ess which is often conspicuous at the tip of the primary receptacle (Plate XLI, figs. 10, 14, 18). The signifi- 

 cance of this appendage I am unable to suggest. Although very characteristic, it may sometimes be absent. 



