284 PROCEEDINGS OF SECTION D. 



Josh. (I.e., PL iv, f. 5) ; Doc. trahecitla var. Delpontei, f. mediolcevis, 

 {PI. mediolcBve), I.e., PI. ii, f. 10, whieh are senile forms of Pen. 

 cucurhitinum ft subpolymorphum, Nord., Ar. gibber ulus var. ellip- 

 iicns (Playf.), I.e., PI. iv, f. 4 ; and Doc. trabecula var. Delpontei mihi 

 respeetively. The same condition may come about naturally in 

 course of time if the cell survive long enough without division. 

 That this is caused by a feeble life I had demonstrated to me lately. 

 Obtaining a gathering of Doc. trabecula var. crenidatum. I put it 

 up in phials, part with preservative and part in plain w^ater. 

 Examining these samples some months afterwards I found that the 

 majority of the specimens in water had become incrassate, while 

 those in preservative showed the original condition. 



An appearance of incrassation is occasionally noticeable, 

 generally in front, but sometimes in side view. This is an optical 

 illusion caused by close pUcation in the membrane, which produces 

 an appearance of extra solidity through refraction. The same 

 appearance may be observed when the grooves in the bottom of 

 cut-glass dish, or tumbler, are viewed from within. A similar 

 refraction in front view is produced by vertical plicae in Cos. 

 rectangulare var. sulcatum, by papillae in Cos. Bceckii var. trifoliattini 

 (Playf.), by the inflations of many species of Euastrmn when the 

 cells are put a little out of focus, and by the prominent tumour of 

 St. assurgens. 



Mixed Forms. — The bulk of the evidence regarding the 

 identity of various " species " and the connection existing between 

 the variations must necessarily be derived from those cells, only 

 too sparingly found, imfortunately, in which the semi-cells belong 

 to different types. And the fact of these occurring at all is a proof 

 (1) that one form can develop into another, and (2) of the powerful 

 effect exercised on development by a sudden change of conditions. 

 These mixed forms must be produced at division, and the rule 

 among Desmids is that the young semi-cell develops immediately 

 into the facsimile of the other, however degenerate the latter may 

 be. But in a mixed form, in the midst of this process some change 

 of environment has been experienced, making either for a stronger 

 or weaker growth, and the young semi-cell attains to a higher or 

 more probably falls to a lower form than that of its mate. 



In such mixed forms, while division is proceeding, each semi- 

 cell retains its own shape, for it is a law, or rather a necessity, of 

 the case that development of the semi- cell cannot proceed while 

 cell-division is going on. A mixed form, A+B, therefore, at its 

 next division produces two cells, A+A and B-rB, andin this way 

 all sorts of variations are created and multiplied. It should be 

 noted, as some explanation of the spread of variations, that a 

 mixed form develops in equal proportions into the type and the 

 variant. Such a variation B+B has a well-defined shape, but 

 it is not on that account permanent. Let circumstances come in 

 favourable to cell- development, and not to division, and B-|-B will 

 immediately develop back again into A+Aor into some still higher 

 form. 



