ON HOLOSCHCENUS LINK. 823 



"deorsum scabrse"; the smooth three filaments are also shown. 

 In the diagnostic specific character of one of the species Reichenbach 

 pives " nuce setas aequante " ; and in the descriptions of each of the 

 three plates he says he depicts " nux cum setis." 



Recollecting the minute and long attention that Reichenbach 

 had paid to Holoschcenus, I think these statements with figures of the 

 setae are the most extraordinary blunder I can call to mind in the 

 whole extent of systematic botany. I have no doubt that the setoe 

 are invariably and absolutely wanting ; I have never found a seta 

 even one-tenth the length of the nut in any specimen of Holoschcenus^ 

 and I have examined many of the ipsisdma examples of Reichenbach. 

 It is impossible to suppose that Reichenbach did not know a seta 

 from a filament : his figures show that he did ; moreover, he must 

 have made some examination, for he distinguishes one of his (very 

 bad) species by its " nuce setas aequante," whereas in his other two 

 species the nut was shorter than the setae. 



Botanists subsequent to Beichenbach's Ic. FL Germ, have been 

 able to make nothing of the puzzle. No one of them (unless the 

 observation has escaped me) anywhere says that he has found (ever 

 so small) a seta in any " species " of Holoschcenus. Parlatore (FL 

 Ital. ii.) says, M Setae nullae," but cites Beichenbach's figures 

 without comment. This is remarkable, as Parlatore is exhaustive 

 almost to tediousness ; his silence I interpret to mean that he did 

 not know what to say. Bos well Syme gives " Hypogynous bristles 

 none (' 4 to 6,' Reich.). " Grenier et Godron, Sir J. D. Hooker, 

 simply say, " Setae 0." Cosson et Durieu, Explor. Alger, ii. 



p. 236, say, "Setae nullae," but cite Reichenbach's figure without 

 comment. 



The only explanation that I have been able to conceive is that 

 Reichenbach had his plates of Holoschcenus done some time, perhaps 

 years, before they were published ; that his artist was a much less 

 able botanist than himself ; that when subsequently Reichenbach 

 "went to press" he was hurried, and worked from the plates, for- 

 getting that the analyses were not, or might not have been, his own. 

 To sum up, no trace of a seta is (as I believe) ever to be found 

 in S. Holoschcenus Linn. But there is an example of it, in the 

 Calcutta Herbarium, believed to have been collected in the West 

 Himalayas or North Cabul (but the habitat rather dubious), in 

 which there are two (nearly lateral) squatnae, longer than the nut, 

 one broader than the other. These are doubtless homologous with 

 the two squamae whose (occasional) presence alone separates the 

 genus Hemicarpha (a bad genus) from Scirpus. These two squamae 

 (once seen) in S. Holoschcenus would doubtless be considered by 

 Bentham (as the similar and similarly placed two squamae in Scirpus 

 membranaceus , once seen) as bracteoles (vorblaetter). I do not so 

 regard them ; I believe they are homologous with the setas (as I 

 attempted to show the Linnean Society two years ago). But, 

 whatever they are, they are totally unlike the setae figured by 

 Reichenbach in his three plates of Holoschcenus, nor could Reichen- 

 bach ever have seen the Calcutta specimen, which alone possesses 

 these "lateral squamae. " 



y 2 



