1 T^ 



498 MR. H. N. DIXON: THE MOSSES OF THE 



CAMPTLorrs STBCOMOSUS, Dixon, sp. nov. 



(Eu-campfjlopvs^ Atrichia) Gracilis^ liumiliSj ca^spitom densuiiij vix 1 ctn. 

 altnm, olivaceun}^ inferne nigrltuiu fornians. Folia- falcato-docurva, sicca 

 valde Jlexaosiiy (j-8 mm. longa, curvato-floxuosaj e basi brevi oblongo-ovata 

 paullo roncaca senaiiu in subulani lotigani graoiloiu concavo-convolntain 

 angustata, niargini1)iis e medio folio luiiiuto conferte Jcnticulatis, superno 

 dense^ sat avijule dentatls, Costa lata, circa Jlmidiain partem folii lutitiulinis 

 a^quan^:, supenic partem maximum ?ubula3 occupaiis^ dorso spieuluso-denti' 

 cidata ; sectioiie colliilas ventralcs, maxhnas^ inancs, duces parvos, cellulas 

 dorsales minutas cum .stcrcideis intormixtas cxliibenSj dorso baud lamellosus. 

 CellulrG alares fusco-purpureo?, mtionnOj laxa? ; senshn in supra-alarcs laxas, 



roctangalares (circa 2-3x1) incrassatas traiiseuntcs, marginalibus sonsim 

 angn^tioribns, omrvhus cJd(rrojdij/Uosis^ supcriores broYiter rbouiboidoo-rect- 



singularcs, serie una siibula? m;irgin:di plerunique inanes, hyalina?. Cetera 



ignota. 



(No. 1.) Tbis appears to bo closely allied to C* comosffs (Hornsch. tt 



Iteinw.), Bry. Jav., in the structure aud areolation oE the lower part of tlio 

 leaves, notably In the absence of any kind of supra-alar narrow liyalino cells, 

 the areolation beinor in fuct subriinular tbrou^iliout, only becomint)- o-raduall\* 



^ J — J ^ ^ 



smaller upwards and outwards, and passing gradually into the enlarged alar 

 cells^ which therefore do not form any ^voll-defined auricles. The subula, 

 morcovor, is much longer and finer, and is closely denticulate throughout 

 nearl}^ its entire lengthy sharply so at apex, and is also scabrous at bad:. 

 The whole plant is much shorter and more slender than in C. comosus^\\\nchy 

 indeedj it does not resemble at all closely in habit. 



LEUCOBRYACEJ^.. 



Lelcobuyuji sanctum (Brid*), C. Muell. Syn. i. 77 *. 

 (Xos- 8, 10.) 



Leucobryum candidum (Brid.), var. srEiuosTicnuM [G. MuelL), Dixon, 



-comb. nov. {^. sjjeirosiicltum^ C.Muelb, in sched,) 



(No, 7.) L. randidum has not I believe been recorded from New Guinea. 

 'The var. spelrostlchum is in its best-developed forms a striking plant ; but 

 all IntGrmediates can be found in the New Zealand forms; both the smaller 

 forms and the var. ntajus give rise to it. The smaller form corresponds to 

 L. Tfij^smannianum^ Bry, Jav,, and the larger to L. pentastichim^l^vy. Jav,; 



* i. sanctum is cited by all the authorities I know of— C. MuelL Syn. ; Bry. Jav. ; Paris, 

 Ind.; Fleischer ; Brotlicms — as L. sanctum^ Ilampej in Liruia^nj xiii, (1839) 42, Thei'e is jio 

 reference to the Javanese t^pecies there, howeverj and C. Mueller's in the Synopsis appears 

 to be the earliest combination. 



