7 



it has seemed better tu take up Lamarck's earlier name, to Avliich 

 ii short description is attacUed. ... 



Tlie true T. paniculatum of Linnaeus proves to be identical 

 with T. tenuc, IJernhardi (Flora, 1845, p. 81), a plant belonging 

 to the section Barhata and quite distinct from T. virgatuin. 



T. dehiUy Sj^reng., from the specimen preserved at Stockholm 

 ■tilso ai-»pears to be a weak straggling form of T. paniculnfuwy 

 Linn. 



The confusion over i\\e name T. pan icula turn is furtlier 

 iacreased by tlie fact that tbere is a specimen with tin's name in 

 the Thunborg Herbarium (.s-re Thunb. Fl. Cap. oJ. Schult. p. 210;, 

 <lifforent from any of the specimens to Avhich reference has been 

 made. This plant was made the type of T, Thunhergianum by 

 De Candolle (Prodr. xiv. p. 6GC), but as Sondta- very riglitly points 

 •out, T. scJayineiun, A.DC. (Esp. Nouv. Thes. 3; Prodr. xiv. G58) 

 is the same thing as Thunberg's plant and has been wrongly 

 daced by Dc Candolle among the unbearded species. T, panicu- 

 afHWy Thunb., and T. Thunhcrgianum^ A.DC, have therefore 

 l)een reduced to T. .^eJagineuiUy A.DC, as it has not been possible 

 :to retain Tliunberg's name. 



One other point desej ving of notice is that certain 8j>ecies appear 

 ^o be vanisliing or may already have been lost; T, rigichnn^ T. 

 Mlirersifoliinn, 1\ wacrostnrhipnn and T. micropogon, for instance, 

 Avere all collected by Ecklon and Zeyher, but do not appear to 

 have been found since by any other collectors. It niay be tliat 

 they have been entirely destroyed by bush fires. BuroIieiFs plants 

 ill many cases are also represented by his ^specimens alone, but 

 iliis is probably to be accounted for by the fact that many of his 

 •collecting grounds have not been revisited. 



In drawing up a hey to the species a good deal of use has been 



inade of the different types of inflorescence, since on this character 



•affinities appear to be more clearly marked for the smaller groups 



of species than Avith any otlier. At times, especially with tbe 



looser types of flower arrangement, it is not easy to draw a bard- 



and-fast distinciitm betwc^en loosp cymo*«e racemes and panicles, but 



for the general broad types the distinctions seem fairly sound. In 



the case of the plants with solitary or subsolitarv flowers and leafv 



stems belonging to the bearded section (7", rariflorum^ T, Zeylieriy 



T. cytisoidesy i\ Tiitrchellii), no doubt a verj' artificial grouj^ lias 



resulted, siiice the plants collected togetlier probably have their 



fillies among species with uureduced flower lieads, m hilst the other 



group with solitary flowers, including T, scrtiilaria.^frum, T. 



panicuJatuviy L. (T. tenue Bcrnh.), T. euphrasioules, T, micro- 



mciiay T. capitjiViforitin and T. cuspidaffnu appear to forui a 



rgroup of closely allied siiecies. 



The treatment of some of the spinous plants is also no doubt 

 5;oTnewhat artificial, but for tlie sako of simplicity in drawing up 

 ilie key it seems better to retain a few inconsistencies; in urr-niire- 

 went. 



In revising the genus for the Flora Copensi.^ it has l)een found 

 liecessary not only to redraft the generic descrlpticm, but also in 

 •order to prevent confusion to redefine Ihe sections and give tliem 



iresh names. 



