SOL'Ti: Al-UUAN 1 1 ICl'ATKJ:. 42/ 



forests, combined willi intense sunshine during summer where 

 not sjiaded, naturally exhibit extremes of colour and of vigoui- 

 less frecjuent in some other lands. A s})ecies may be dark brown 

 or almost Itlacis on one side of a >tone or tree, and Iv/ight green 

 or glaucous on the otlier side, with every gradation between, 

 hence the futility of tru>ting to ci'>lour as a specific character, 

 except in a general wa} . A s|)ecies. where loosely attached to 

 a stone or stumjj, may develop a depauperate condition very 

 unlike an adjoming \igorons patcli : or an exten>ion of the 

 patch iiito dense shade, ^uch as under a stone. ma\- become 

 weak, etiolated, sparse-leaved and flagelliferous ; hence the neces- 

 sity of knowing each ]>lant in nature, under its varying circum- 

 stances, rather than accepting a single herbarium scrap as a 

 specific type. 



The variation of leaf-form, arrangement, margin, texture 

 and size, and also of the folioles. tioral leaves and perianth, even 

 ui)on a single stem, are often niost confusing, and emphasise 

 the necessity of studying many specimens in a tuft l)efore arriv- 

 ing at a conclusion as to the general t\'pe-form and th.e range 

 of its variation, one of the greatest difficulties being that related 

 species frequentl}- range more or less in the same direction, 

 though normall)- ])osses-ed of characters which render tiiem 

 f(tiite distinct. 



And perhai^s one of the greatest difficulties to tlie Ijeginner 

 is the frequent intermixture of s])ecies. often clo>el\' related, 

 which has to be very carefullv guarded against in order to avoid 

 confusion. 



Some of the thalloid s])ecies are annual or u>uall\- annual. 

 others are perennial, and in the fcdiose group ] am not aware of 

 an\- anntial species under conditions suitable for ])rolonged life, 

 but many in that grou]j continue to gro^^• on b\' means of ter- 

 minal or lateral innovations, while the older portion^ d.ie away, 

 and in certain genera it i- tlie haliit for mature stems to become 

 ])r()strate (resembling rhizomes), and ro i)roduce adventitious 

 innovations wdiich become new stem*, therebA' perj^etuating the 

 grcnvth from time to time and forming loose ctishions. 



No true roots occur, but long, one-celled rhizoid- take their 

 ])lace, sometimes (among thalloid species) of two distinct kinds, 

 the one lar^-e and oi^en. like a vessel in i^'O'-enchvmatous tissue, 

 antl evidentl)' intended fr)r the free flow of sap in quantity, and 

 the other more slender and ])rovideil \\ ith Avartv excrescences on 

 the inner surface, whicli formatiori is ex])laincd to be intended, 

 or at least to act, as an a^->istance in \\ater-carriage where air is 

 ])resent in the cell, which might otherwise form a fttll-sized 

 bubble and stop the ])assage of water, the excrescences holding 

 the bubble in the centre of the lube and allowing water to ]iass 

 alongside. 



Another explanation is that the projecting pog> lead to an 

 increase on the total area of the ectoi^lastic mcmln'ane which 

 lines the protoplasm of the cell inside the cellwall. and is known 



