Hi 



822 ON HOLOSl'HlENUS LINK, 



bach admitted the three old Linnean species (under new names), 

 published two more "intermediate" species, and issued under MS. 

 names one or two more species. The whole of these are reduced 

 to S. IMoschcenus Linn, by Boeckeler — I think rightly. So also 

 Boswell Syme ; the only English locality known to me is Braunton 

 Burrows, in Devon, where the plant is 8. " 



typical. 



It is not the object of this paper to criticise " species " or " sub- 

 species" of Holoschamus in detail. The case is one that occurs in 

 perhaps one-third of our species of well-known wild plants. We 

 have first the standard plant; then we have the S. Romanus Linn, 

 smaller, dried up, with few small spikes on short rays ; then we 

 have S. globiferus Linn. f. (also a southern form), with a compound 

 eyolute inflorescence, but smallish spikes (sometimes as many as 

 eighty). Under these three forms, perhaps three-fourths of the 

 material in the large herbaria can be sorted out. The remaining 

 one-fourth provides an almost boundless scope for the invention of 

 new " species " ; or " hybrids," in the more modern style. I would 

 once more protest against such plants being recorded as "hybrids." 

 First, because there is no experimental proof that they are hybrids, 

 or from what parents they are derived ; I do not see why many of 

 them may not be "varieties" (possibly even from seed in a self- 

 fertilised flower), due to soil, climate, &c. Secondly, because, within 

 the range of variation within which fertilisation is nearly perfect, the 

 results of intercrossing between varieties and "subspecies" should 

 be called only crosses. The study of " hybrids " in the Darwinian 

 sense between well-separated species which (usually) hybridise only 

 with difficulty has led to great results; and experiments in this 

 sort of hybridism appear to me much more likely to prove valuable 

 than labours in the fine separation and distinguishing of species. 

 But I doubt if observations or even experiments in botanic inter- 

 crossing of very closely allied (dubious) wild species are very promising 

 of theoretic results ; so much has been done in this direction by 

 gardeners. ■ I do not think it an advisable use of the term to call a 

 Halt- bred Lincoln sheep a hybrid, though we actually know in this 

 case that it was a cross between two distinguishable subspecies. 



in systematising, I would certainly follow Boeckeler and others 

 in treating all this Holoschcenm as one species, *. e., I think it would 

 be a waste of time for any botanist who attempts general work to 

 trouble himself with the forms of " Romamts," " globi ferns," &c, 

 and tlieir bewildering series of connecting forms ; I do not say that 

 irom economic or other causes it may not be worth while for some- 

 body to study the group. 



♦^q^SS *' in Ic ' FL Gen "- et Helv - ^ ( 1846 )> PP- 44 > 45 > 

 tabb. 316-818, hgg. 736-741, again takes up Holoschmnus, admitting 



thiee species in the German Flora. He now gives, as a character 

 .?,!, %wl dia S*osis of the genus), " Setae hypogyn® 4-6." In 

 each of his three tabb. he gives analysis of the flower ; in tab. 316, 



1™ S1X f t£e ^ oufc twice as lon S ^ the nut ; in tab. 317, he 

 fonr Ifl « 8 ? rat 16r exce ^ing the nut ; in tab. 318, he shows 

 four setae nearly as long as the nut. These " seta " are all linear 



