29G 



fore rac'liifonn^ bear the sori terminally^ so that eacli pinnule 

 looks like the stalk of a soriis. The branching of the fertile 

 pinnne, though similar in degree to that of the sterile pinnae, 

 is some^-hat contracted, and departs from the horizontal plane 

 in such a way as to assume a thyrsoid character, to which the 

 name of the genus refers. 



A cup-like basal indusium invests the sorus, which consists of 

 numerous sporangia covering a prominent receptacle. The 

 sporangium has a continuous, oblique, irregular annulus with no 

 definite stomium. The sporangia are developed on the receptacle 

 in basipetal order. 



The genus Thyrsopteris was established in 1834 by Kunze* to 

 receive this species, which till then had not been described, 

 though specimens were collected by Bertero in 1830. Tltyrsopteris 



t 



faInilv^ In H 



Species riliciini (1846) Tliyrsopteris was placed next to 

 picksonia, but in tlie Synopsis Filicum (Hooker and Baker, 18G8) 

 it was transferred to a position next to Cyatliea, wliile Dicls (in 

 Engler and Prantl, jSTatiirl. Pflanzenfam.) inakes a tribe {TJiyrsop- 

 terideae) for it between DlcJcsonieae and Cyatheae in the 

 Cyatlieaccae. BowerJ, wko Las studied tlie development of tlie 

 sorus and sporangium, liolds tliat tlie cliaracters of Tliyrsopteris 

 indicate a relationsliip to Dichsonia, tlioiTgh not a very close one, 

 and favours the view rbat tbe Tliyrsopterideae should rank as a 

 separate family. The indusium at an early stage is slightly two- 

 lijpped, which is an indication of correspondence with DicJcsonia- 

 {Cihotiuvi). A somewhat archaic character is recognised in 

 certain structural features of the annulus. 



The occurrence of fertile and sterile tracts of very differenir 

 appearance on the same leaf, though a rather rare phenomenon, 

 is not restricted to a small phyletic series of Ferns, but is met" 

 with again in genera far removed from Thyrsopteris, viz. : 

 Osvninda and Aneimia. 



Thyrsopteris is a monotypic genus as regards living plants. 

 Among fossil plants some Ferns of Jurassic, Wealden and 

 Cretaceous age have been referred to this genus by certain authors, 

 bnt it appears that the data obtained were not sufficient in anv of 

 these cases to warrant so precise an identification^. 



Ferns form an important constituent of the Flora of Juan 

 iernandez, the species of Ferns recorded (45) amounting to 

 nearly a third of the vascular plants. Of the Ferns, six besides 



frond of which is 



the Thyrsopteris elegans are endemic species. 

 The larg ' • - ~ 



illustrated in me accompanying plate, has been at Kew for many 

 years, and its exact history is not on record. It thrives in a 

 mixtiire of peat and loam placed between rough stones and enioys 

 a position where it obtains plenty of moisture and some shade. 



* Kunze, Linnaea., vol. ix, p. 507. 



PotinsiSL?.^?^^'^ "" "^^' to his spocimen.s as iolh^rs :-" Cyalhea? an 

 puuaa j^neimeae i^pecies nova n. 1537 



I Bower, The Origin of a Land Flora, 1908, o. 589 

 S See Reward, Foss Plninf« ,-r.i o J ',ao '/ ^^- 



