374 MONOGRAPH OF TIIE LABOULBENIACEiE. 



pressure of the spore mass, and there are indications that they are absorbed even before this 

 pressure is exerted. 



The procarpe, as we have seen, is identical with that of other genera, and the further 

 development of the carpogenic cell, although it has not been traced out in detail, seems to corre- 

 spond in all respects to that which has been previously described. In species in which it has 

 been possible to determine the number of ascogenic cells with defmiteness, four have been seen 

 in every case ; but a certain irregularity in the position of the asci in relation to them is often 

 observable, the ordinary biseriate arrangement being at least obscured. 



The young individuals of species of this genus bear a close resemblance to those of Zodio- 

 myces, and the slender acicular spores, as in the last named genus, are septate nearer to the base 

 than to the apex except in C. terrestris. A further peculiarity of the spores is observable in (J. 

 furcatus and C. contortus, in both of which the septum that divides the spore involves its gela- 

 tinous envelope also, which is prominently constricted in this region (Plate XXV, figs. 4 

 and 10). 



The trichogyne is always filamentous, usually not very long, and simple or sparingly 

 branched. The antherozoids are clearly differentiated only in C. rostratus (Plate XXIV, figs. 21 

 and 24) and C. terrestris (Plate XXV, fig. 23), in both of which they are long, rod-shaped 

 bodies that fall from their attachment soon after they are formed. In other species they appear 

 to result from the segmentation of slender branches into rod-like pieces like those of the two 

 forms mentioned. It is difficult to obtain trichogyncs in a receptive condition, and in only a 

 single instance have I found one to which three or four of the long antherozoids were firmly 

 attached. 



The species are in a way divisible into sub-groups corresponding to the genera of the hosts 

 which they inhabit. The three species on Berosus as well as the four on Tropisternus have a 

 distinct family resemblance in either case ; while 0. rostratus is almost sufficiently distinct to be 

 placed in a genus by itself; its chief peculiarity being connected with the production of adven- 

 titious branches from about the base of the appendage and perithecium, and the definite and 

 successive production of its antherozoids. It may be mentioned that it is approached in both 

 these respects by the singular little terrestrial form C. terrestris. What the near affinities of 

 this peculiar and comparatively imperfectly known genus are, it is difficult to say, although its 

 exogenous antherozoids and the form of its young conditions indicate a nearer relationship with 

 Zodiomyces than with any other known forms. 



The species all inhabit aquatic beetles of a single family (Hydrophilidae), with the exception 

 of 0. terrestris, which is found on a staphylinid that lives in very wet situations. 



Ceratomyces mibabilis Thaxter. Plate XXIV, figs. 1-10. 



Proc. Am. Acad. Arts and Sci. Vol. XXVII, p. 34 ; 1. c. Vol. XXX, p. 480. 



At first pale, becoming more or less deeply suffused with amber-brown. Perithecium straight 

 externally, more or less inflated internally, the cell-rows composed of from twenty to thirty cells, 

 the anterior row ending below the apex in an appendage often equalling the perithecium in 

 length and tapering to a blunt point, or when broken sometimes producing a few slender termi- 

 nal branches, becoming blackened and inflated near its base, composed of from ten to sixteen 



