24 PEOCEEDINGS OF THE 



Pig. 3 shows the case where the nut is intermediate between 

 triangular and flattened ; 2 style-branches are perfectly deve- 

 loped, 1 very imperfectly, viz. that corresponding to the flattened 

 anticous angle, which is reduced to a tail without stigmatic papillae. 

 The wing, fully developed on the other two angles, is very small 

 and imperfect on the anticous angle. All stages in the reduc- 

 tion from triangular to flat can be illustrated from this species. 



Throughout the Order almost this reduction is frequent, as in 

 Cyperus, Uleocharis, Scirpus, &c. "We may suppose the cause to 

 be the pressure of the stiff keel of the glume : when the reduction 

 is " settled^'' the glume is usually round-backed. In the large 

 number of species, the reduction either has or has not taken place ; 

 so that the character taken from " style 2-fid, style 3-fi.d " is a valu- 

 able one to systematists. But there are various species, and as it 

 happens several European species, in which the plant has divided 

 itself into two " subspecies," whereof (a) mostly follows the 

 antique, symmetric form ; the other (/3) affects strongly the 

 modern, more specialized form. I may cite, as you all know it, the 

 common Bulrush, Scirpus lacusfris, Linn. This has usually the 

 style with three equal branches. But there is another state of it, 

 tlie Scirpus Taherncemontani, which has the style 2-fid. Collectors 

 and systematists know well the disputes over the distinctness 

 of these, and the synonymy which encumbers a question really 

 of terms. It will satisfy an evolutionist to see that he has in 

 these cases the formation of a new species going on before his 

 eyes: there was perhaps, not long ago geologically, only one 

 form ; and there will be perhaps two very good species another 

 day. 



I note that, though these transitions from 3 to 2 style- 

 branches by reason of the compression of the front angle of 

 the nut occur in many genera, as Cyperus, Scirpus, Eleocharis, 

 the percentage of occurrence is much smaller than moslfEuropean 

 botanists suppose ; because several transitional cases happen to 

 occur in common European plants. Transitional cases hardly 

 occur in Carex, except in the very common European Carex glauca 

 (and a few others). In these the lowest, strongest, flower has 

 nearly always a 3-branched style. 



I now come to the much more remarkable case where the nut 

 is compressed laterally in the words of Kunth and Nees, as re- 

 presented in fig. 4. In this reduced form (genus Fycreus) which 

 occurs, for instance, in one-sixth of the species of the old genus 

 Cyperus, the plane comprising the two style-branches passes 

 also through the axis, and they lean away unsym metrically 

 from the axis. 



Now I have never been able to discover the slightest trace of 

 the route by which this reduction took place. I have never even, 

 as a monstrosity, known any one of the genus Pycreus (or other 

 genus with laterally compressed nut) to show any trace of a 

 third style, I know only too well that there are many species 



