286 C. SKOTTSBERG 



II. Neotropical element. — 26 sp. (21.7 %). 



a. Endemic species (11): Campylopus subareodictyon, Leptodontium fernan- 

 dezianum, Didymodon calymperidictyon and linearis, Bryuni fernandezianum, Pin- 

 natella macrosticta, Isopterygium fernandezianum, Rhaphidostegium Masafuerae 

 and caespitosoides, Rigodium robustum and Looseri. 



b. Also in S. America (15): Gymnostomum calcareum (widely dispersed), Cam- 

 pylopus areodictyon, Thysanomitrium Richardi, Mielichhoferia longiseta, Anacolia 

 subsessilis, Porothamnium fasciculatum and arbusculans, Lamprophyllum splendi- 

 dissimum, Rhacopilum fernandezianum, Stereodon Lechleri, Rhaphidostegium 

 caespitosum, Rigodium toxarium, arborescens, hylocomioides and tamarix. 



III. Chilean element. — 24 sp. (20.0 %). 



a. Endemic species (4): Fissidens pycnotylus, Ulota fernandeziana, Philonotis 

 glabrata, Rhaphidostegium aberrans. 



/;. Also in Chile, mostly southern (17): Pleuridium Robinsonii, Hymenosto- 

 mum kunzeanum, Oncophorus fuegianus, Campylopus truncatus, Fissidens lepto- 

 chaete and maschalanthus, Tortula scabrinervis and flagellaris, Rhacomitrium lori- 

 forme and convolutam, Stenomitrium pentastichum, Bryum Lechleri, Bartramia 

 aristata, Philonotis krauseana and vagans, Rhynchostegium complanum, Oligotrichum 

 canaliculatum. 



c. Cosmopolitan (3): Ceratodon purpureus, Rhacomitrium lanuginosum, Funa- 

 ria hygrometrica. 



A^ote. — On p. 234 Campylopus iiilroflexus was placed here; it was transferred to 

 I: I on account of its circumpolar distribution in the far south, from where it may have 

 invaded the tropics and migrated north, but it is perhaps just as probable that it is 

 an old pantropical species which has spread both north and south. 



IV. Atlantic-Mediterranean element. — 3 sp. (2.5 %). 



Comp. above, p. 235. 



a. Endemic: Fissidens crassicuspes. 



b. Not endemic: Campylopus polytrichoides, Trichostomum brachydontium. 



We have seen above that Herzog regarded the moss flora of Juan Fernan- 

 dez as almost entirely austral-antarctic and that the tropical element was small. 

 Most likely he added the species forming my Chilean element, because they are 

 concentrated in the moist southern part of the country and in several cases range 

 south into the subantarctic zone which is, however, no proof of their Antarctic 

 ancestry. A species like Porothainnmni arbusculans is, as far as I know, restricted 

 to Chile and Patagonia, but the genus is essentially tropical, Stereodon Lechleri 

 extends from S. Chile to W. Patagonia, but is the only species reported south 

 of the Equator; Rigodium is American except 2 species found in Africa and 

 essentially tropical. On the other hand it is difficult to draw a limit between 

 groups II and III, but the species with a southern area are so numerous that I 

 felt obliged to make this distinction. The Atlantic element will perhaps disappear 

 when the distribution becomes better known. 



