I 



ANALYSIS OF Till: WoKK OF CARL TIlUNBKKG. ,^8l 



the beautiful Scrniria florida Kn., which he described and 

 fi.qured as Profca florida. 'riii> rare species was lost sight 

 of until rediscovered l)y the late Prof. M;ic(>\\an in the 

 same locality as Thunberg- gathered it. LAilike otlier botanical 

 collectors such as Burchell, Drege, Ecklon. and Ze\her, 'iduuiberg 

 recorded few localities, and neitlier from his ])ul)lications nor 

 from his herbarium is it possible to disco\er exactly wliere he 

 gathered many of his Protcacca\ Out of the 79 species which 

 he collected we find only 16 with the localities given. 



The examination of T]ninl)erg"s herljarium was of the utmost 

 value in rectifying many errors of synonymy in Aleisner's* mono- 

 gra])h, which otherwise would still have remained doubtful, as 

 evidently Aleisner had not access to the collection. The exam- 

 ination of his specimens in connection with his publications was 

 the only way of determining the sj^ecies to which he refers in 

 his works, and though Thunberg's descriptions are too meagre 

 to be used as a means of identifying a species, they were very 

 accurate of the specimens he described. Thunberg's lirst pub- 

 lication t on the Froleacccc appeared in 1781, and was a consider- 

 able advance on the literature of the day dealing with this group. 

 He gives a large amount of useful informaticn about the habits 

 of the species he collected, and described 60 species under the 

 genus Protea, and illustrated the text with 5 plates, and for the 

 first time figured species of Spatalla and Soroccplialiis. In 1794 

 he published the " Prodromus Plantarum Capensium," and men- 

 tioned 61 species. It was not until 19 years afterwards that he 

 published his first editii n of the " Flora Ca]:»ensis," followed 

 in 181S by the second edition of the same work. The whole of 

 the work was ultimately edited from the author's manuscript 

 in 1823 l)y Dr. A. Schultes. at Stuttgart. Thunberg's delay 

 in publishing the results of his collection was unfortunate, as 

 bv the time his " Flora Capensis "' api)eared it was out of date 

 as far as the information about the Proicacdc was concerned, 

 as in 1809 KnightJ described 190 species, and a year later Robert 

 Brown§ published a paper which contained descriptions of 181 

 species. Thunberg only mentions 80 species in the " Flora 

 Capensis." In all his writings Thunberg described 84 species 

 of Proteacccc, two of which, however, were not South African, 

 viz., Protea hirta Thunb., Fl. Cap. i. 454, is a species of Isopogoii 

 (I. aiictliifolius Kn.), collected in New Holland, and Protea 

 coarctata Thunb., Fl. Cap., ed. Schultes, 122, is Petropliila pul- 

 chclla R. Br., an Australian plant. Points such as these it was 

 only possible to clear up by reference to his herbarium. -Vgain, 

 in cases where Thunberg lumped several species under one name 



* D.C. Prodr. XIV. 

 t Botanico dc Protea. 



\ " The cultivation of the plants bclongin;^- to the Natural Order 

 i'rotea.' 



§ Trans. Linn. Soc. U^iu), X-. 



