88 FLORA ANTAECTICA. {Auckland and 



snrmnit. This thickening of the three sides takes place during the growth of the seed-vessel, as in the young ova- 

 rium the sides are much narrower than the apex, which is convex and hispid, and which answers to the convex 

 base of the hollow at the top of the ripe nut. I have examined several embryos in all the species, and very many 

 of the present ; their structure and forrn are exceedingly constant, lying in the very bottom of the seed, the broad 

 upper end sunk in a shallow fossa at the base of the albumen. 



Plate XLIX. Fig. 1, a leaf; fig. 2, a flower enclosed in the glumes; fig. 3, the same, with the glumes re- 

 moved; fig. 4, a stamen; fig. 5, ovarium, style and stigmata; fig. 6, immature nut after the style has fallen away; 

 fig. 7, a ripe nut enclosed in the persistent perianth ; fig. 8, longitudinal section of a nut, showing the seed ; fig. 9. 

 a seed removed, cut vertically, showing the embryo : — all magnified. 



2. ISOLEPIS, Br. 



1. Isolepis AucJdandica, Hook, fil.j pusilla, dense ceespitosa, culrnis erectis setaceis basi divisis foliosis 

 teretibus polyphyllis, foliis culmo subsequilongis semiteretibus supeme canaliculatis striatis, spica solitaria 

 lateralis squamis paucis omnibus floriferis, staminibus stigmatibusque 3, nucibus elliptico-ovatis trigonis 

 laevibus pallide straniineis. (Tab. L.) 



Hab. Lord Auckland's group and Campbell's Island; in moist places especially near the sea, also 

 amongst grass on the lulls, abundant. 



Radices caespites 2-3 unc. latos formantes, fibrosae, fibris elongatis, creberrime intertextis, tortuosis, fibrillosis, 

 atro-fuscis. Culmi plurimi, dense fastigiati ; basi crassitie pennae passerinse, subelongati, i unc. longi, nodosi, ad 

 nodos fibrillosi, vaginis rufo-castaneis foliorura vetustorum obtecti, ter quaterve clivisi ; supeme subvalidi, nudi, 

 erecti, 24— 4-unciales, teretes, striata. Folia 2-6, plerumque 3-4, erecta, basi vaginantia, subcoriacea, glaberrima, 

 laevia, filifonnia, obtusa v. apice rotundata, culinmn plerumque paulo superantia, interdum i- J unc. lata, semiteretia, 

 dorso convexa, supra canabcidata, marginibus subinvolutis v. planiusculis, sub lente bneis alternantibus viridibus 

 albidisque striata. Vagina 4/ unc. longae, teretiusculae, eompressae, basi rufo-castaneae, nitidae, nervosa?, amice 

 membranacese, reticulata?, ore oblique truncato, intcgeirimo, ligula nulla. Spicula solitaria, v. rarius spieidae 2, parva, 

 linea vix longior, latiuscida, apice truncata, nempe squamis inferioribus elongatis spicam a?quantibus. Squama 

 paucse, 6-8, late ovatae, valde concavae, coriaceae, ad margines late membranacese, in apicem crassum productse, dorso 

 superne subincrassatae, carinatae, lateribus tenuiter 3-5 nerviis, virides v. castaneo purpureove pictae, nitidae. Sta- 

 mina 3 ; filamenta Mnearia, plana, reticulata, superne latiora ; anthera basifixas, elongatae, loculis basi apiceque 

 divaricatis. Ovarium minimum, ovatum, in stylum rectum desinens. Stigmata 3, exserta, elongata, liispida. Nux 

 squama paido brevior, elliptica, utrinque acuta, trigona, compressa, angidis obtusis, glaberrima, lasvis, non polita, 

 pallide flava v. straminea. 



It is not before the most careful examination and comparison of this with many other similar species from 

 various parts of the world, that I have decided upon describing it as new. Nor coidd I make it agree with the 

 description of any of the numerous species of the southern hemisphere. It appears not only to differ from the 

 European, but also from the twelve or fourteen plants belonging to this genus now known to inhabit Australia and 

 New Zealand. In habit and appearance it resembles /. setacea, L. and /. Savii, from which it differs in the leaves 

 being always more numerous and as long or longer than the culm, iu the shorter spikes, and more materially in the 

 seed, which is twice as large as in those species and of quite a different shape, being elhptical-ovate, compressed, 

 trigonous with the angles roimded, the surface is smooth but not shining and the colour pale yellow. My suite of 

 specimens is very extensive, and these characters are constant in them all. The breadth of the leaves is greater in 

 the upland specimens than in those of the sea-coast. 



Plate L. Fig. 1, apex of the culm and spikelet ; fig. 2, a scale and flower; fig. 3, a stamen ; fig. i and fig. 5, 

 ripe achasnia : — all magnified. 



