THAXTER. MONOGRAPH OF TIIK LAHOILMKM ACK.K. 



283 



row. Spores minute, of the usual type, normally discharged in pairs the members of which produce male 

 and female individuals. Asei apparently eight-sporcd ? 



The discovery of (his remarkable genus is due to Mr. Charles Billiard, who first observed it on Erfo- 

 bia. It is unlike all others thus far described from the fact that, in all the species, one or both of the 

 sexes produce secondary receptacles, corresponding to the fertile branches of other genera, which, instead 

 of growing free and depending on the primary foot for their nutrition, attach themselves to the host and 

 become quite independent as far as their food-supply is concerned. Both the primary foot, and the cells 

 of the secondary receptacle, which are in contact with the host, penetrate the latter by means of a simple 

 cylindrical slender rhizoids, readily visible, especially in forms which grow, or begin their growth, as is 

 more often the case, on the usually transparent spines or bristles projecting from the integument of the 

 insect (Plate XXXIX, figs. 3-4, and XLI, fig. S). 



In nearly all cases the original infection appears to take place on one of these bristles, and here, as 

 in other dioecious species, the spores are normally discharged in pairs, the members of which develop side 

 by side (Plate XLI, fig. 17), and grow into male and female individuals respectively (fig. 14). In some 

 species both individuals complete their development on one of these bristles as in //. chwtopliilus, fig. 14; 

 but, in a majority of cases, a fertile branch from the subbasal cell of the primary receptacle of the female, 

 grows downward till it reaches the host's integument, before any production of perithecia begins. More 

 frequently this fertile branch becomes furcate as soon as it leaves the bristle, the two secondary branches 

 being transformed into secondary perithecigerous receptacles, and in several species two paired perithecia 

 are thus produced, one from each of the secondary receptacles, symmetrically placed at the- base of the 

 spine (Plate XXXIX, fig. 9, and Plate XLI, fig. 2). In other cases the secondary receptacles are 

 more highly developed, creeping rather extensively, but never branching, and always growing through 

 the activity of a single terminal cell. Such multicellular secondary receptacles may give rise to numerous 

 perithecia, and this type arrives at its highest development in forms like //. Ectobia> (Plate XXXIX, figs. 

 11 12) and in H. iriruxpidatus (Plate XL, fig. 13). In some forms, also, the secondary receptacles may 

 be more than two, and appear to arise not as branches from the primary fertile branch, but independently. 

 In such cases the subbasal cell of the primary receptacle evidently undergoes secondary divisions, and each 

 of the resultant cells sends down its fertile branch to form a corresponding number of secondary receptacles. 

 I have found it almost impossible to determine this connection by actual observation, but that the secondary 

 receptacles are multiplied in this fashion seems evident from the appearance of detached primary recep- 

 tacles such as are represented in fig. 10, Plate XL, or fig. 12, Plate XLI. 



That still another type exists is indicated by the conditions seen in H. Paranensis and H. Nycto- 

 boras. Although both these species are closely allied to //. tricuspidatus, their secondary receptacles are 

 not continuous cell-series, as in this species (Plate XL, fig. 11), but each perithecium is associated with 

 a shield-like receptacle (Plate XXXIX, fig. 5 and Plate XL, fig. 3). These shield-like structures appear 

 to be quite independent of one another, but that they do not arise, as in the type last described, from 

 distinct branches representing outgrowths from a corresponding cell formed by secondary divisions of 

 the subbasal cell of the primary receptacle, is evident from the conditions seen in isolated primary recep- 

 tacles (Plate XL, fig. 5) from which but a single branch has been developed in the usual fashion. Whether 

 these shield-like structures arise as branchlets from the primary branch, or whether they are to be consid- 

 ered intercalary modifications of it, I have as yet been unable to determine. The former alternative, 

 however, appears to be the more probable. 



A curious condition is also illustrated by H. Periplanetce, which is the most easily obtained anil com- 

 monest member of the genus. In the typical form of this species a remarkable shield- or shell-like struc- 

 ture is developed, that completely covers the secondary receptacles of the female individuals, which are 

 seated on the integument of the host (Plate XLI, figs. 6-7). This shield-like structure is evidently an up- 

 growth from one of the secondary receptacles that compose the group, and is a very characteristic and 

 highly specialized structure. When, however, this species develops wholly on the bristle, and the primary 



