120 ANATOMY AND PHYSIOLOGY OF 



eate filament presents itself, displaying several spiral convolutions, 

 tliickened at one end and running off to a very fine point at the 

 other. Tlie filaments manifest lively motions, exhibiting differences 

 accox'ding to the manner in "which they are rolled up, in some cases 

 while still enclosed in the cells -where they are developed, but 

 more particularly after they have emerged into the water fi^om 

 antheridium, which opens when ripe. Thus, when the filament 

 is rolled up like a watch-spring, the motion is more or less rota- 

 tory, but if it is coiled over in the form of a cork-screw, the move- 

 ment is at the same time an advancing one. In these movements 

 the thin end of the fibre almost always goes first. Minute obser- 

 vation, which in many cases is very difficult, both from the rapi- 

 dity of the motion (which, however, is readily arrested by poisons), 

 and the great delicacy of the whole structure, shews that the 

 movements arise from extremely delicate and comparatively long 

 cili^, of which two are usually found at the thin end of the fila- 

 ment, and which only seem to occur in larger numbers in the 

 Ferns. The filament itself exhibits no independent motion, as 

 indeed, altogether, the land of motion does not indicate any will. 

 The term seminal filaments has been not inaptly applied to these 

 filaments. 



Ohserv, The first observation on the motion of the contents of the 

 antheridia was made by Schmidel ('^ Icones plantarum,^' 1762, 85) in Jun~ 

 germannia pusilla. The imperfection of the microscopes of that period, 

 however, seems to have prevented Ms seeing the seminal filaments, and 

 he probably only observed the cells in which the filaments were enclosed. 

 The same seems to have been the case with the observations made by Fr. 

 Kees von Esenbeck ('^ Flora" 1822, 1, 34) in the antheridia oi Sphagnum. 

 He considered the moving bodies which he saw to be globular monads, 

 and did not doubt their animal nature. The spiral filaments themselves 

 were discovered by linger in the Mosses and Liverworts (" Ann, cl iSc. 

 nak % S&T^ xi. 257) ; in accordance with the then prevalent notions on 

 spermatozoa, he regarded them as animals, and applied to them the 

 name of Spirillvm hryozoon. Recent years have scarcely added to his 

 observations on the seminal filaments, more than the fact that two cilise 

 exist at the thin end of the filameijts, which linger had overlooked (Be- 

 caisne and Thuret, ''Ann, d. So. nat. 3, Ser'' iii 14). Plate 1, figs. 2% — 28. 

 Seminal filaments of Sphagnmn; ■G.g. 26, represents two anther-ceUs with 

 the seminal filaments enclosed ; fig 27, one of the latter seen fi-om the 

 side (from linger). To me the filaments appear to have the form which 

 I have represented in ■B.g. 28, 



The structure of the Moss-anthers is very simple. It consists of a 

 simple sac, with a wall composed of a single layer of cells, which, according 

 to linger, are applied upon the outside of a large cell, while according to 

 ScMmper they are enveloped on their outer sides by a continuous mem- 

 brane composed of intercellular substance. When mature, this coat is 

 torn at the apex, and the contents, now dissolved into a mucilaginous 

 fiuid, issue from it. 



The anthei^s of the Liverworts possess a structure completely analogous 

 to that in the Mosses (Gottsche, ''Act Acad. Nat Cur.^' xx. 1, 293), 



