8 Transactions of the Society. 



family not to mistake for real variations. Sometimes we attain a 

 position where it may be possiblo to simplify this, and to reunite 

 under one species, as merely " forms " of it, various plants, whether of 

 one or of different countries, which their discoverers may have con- 

 sidered to be separate. I believe that this can now be done with the 

 desmidian species Micrasterias americana. 



In a paper of mine in 1880 (Trans. New Zealand Inst., xiii. 

 p. 304)j on New Zealand Desmidieae, I reported the existence of a 

 plant to which I gave the name of 31. ampullacea, and I indicated 

 that it was nearly allied to M. americana. Mr. Archer, in 

 ' Grevillea,' September 1881, referred my plant nearly to M. Her- 

 manniana Keinsch. I understand that Professor Nordstedt, of 

 Lund, would include mine and some others under M. Mahabulesh- 

 tvarensis Hobson. My object in writing now is to advocate that 

 all these, and the other cognate forms, should be merely considered 

 as variations of one type species ; and I select M. americana as the 

 type, because it was the first described. 



The outline of M. americana was very correctly delineated in 

 Mr. Ralfs's work, first under the name M. morsa, afterwards 

 corrected. Since that time, as far as I know, thirteen plants more or 

 less closely resembling the original have been described from various 

 countries. The last of these was reported by Dr. Spencer (Trans. 

 New Zealand Inst., xiv. p. 296 and pi. xxiii.) as a variety of my 

 M. ampullacea, and although at first sight there undoubtedly is 

 no very close resemblance between it and Mr. Ealfs' type, yet when 

 all the fourteen plants are placed together, the gradations are seen to 

 be so gradual that they form a regular series. With the object of 

 showing this, I have attached hereto figures of them all in juxta- 

 position. For most of these figures I am indebted to the kindness 

 of Mr. Barwell Turner. Beginning with the type-species No. 1, 

 it will be seen that the two lateral lobes of each segment are broad at 

 their bases, and are cut at their extremities into four short cylindrico- 

 tapering lobules. ]n the forms 2, 3, 4, and 5, there is not much 

 difference in this respect ; No. 2 has its side lobes apparently even 

 widening towards their ends, or rather with an indication of a small 

 fifth lobule on each side which will be useful for comparison 

 presently. In No. 6 the side lobes are evidently narrower and more 

 deeply incised in the middle, giving an approach to the form No. 7, 

 where the incision is deep enough to produce the effect of only two 

 divaricating lobules. This form passes easily into No. 8, and thence 

 into No. 9, where we have a more pronounced extra lobule than in 

 No. 2. From No. 9 the gradation to No. 14 is quite easy ; in fact, 

 if it were not for other points to be mentioned presently, all these last 

 forms are almost alike. 



In point of fact, judging merely by general outline, the whole 

 series might be divided into two groups : the one including those 

 forms in which the lateral lobes are obscurely bifid ; the other, the 

 forms in which they are distinctly bifid. The extra lobule appears 



