V. RnODOMELACE.E. 31 



length, others not more than as many tenths of an inch ; some dichotomous, others 

 pinnated — some distichous and fern-like, others with a bushy or arborescent cha- 

 racter ; some of cobweb delicacy, lubricous and excessively flaccid, soon decompos- 

 ing, others robust, rigid or tough, of strong enduring substance ; — some of a bril- 

 liant rosy red or crimson, others (and the greater number) varying through all the 

 graver shades of red-brown, brown, and purple ; some inhabiting the deep sea, 

 others occurring only near high water mark or far up the estuaries of tidal rivers. 

 Plants of such varied aspect and habit could not have been brought together by 

 the universal consent of botanists, among whom there has never been much differ- 

 ence of opinion respecting the just limits of this genus, if they had not some 

 obvious bond of union in an essential, easily seen, and important common charac- 

 ter. This is found in the structure of the stem in the articulated species, and of 

 the axis of tlie stem in species which appear to be partially inarticulate ; the dis- 

 sepiments being hid by the growth of a thin or thick layer of epidermal cells round 

 the stem or branches. In such species, the proper structure of the genus may 

 generally be seen, without dissection, in all the younger parts, as in the upper 

 branches and ramuli. The species simplest in structure have the internodes or 

 articulations of the stem and branches composed of four large cells containing endo- 

 chrome or colouring matter, placed crosswise round a small, central, empty cavity, 

 as shown in section at PI. XVI. B. fig. 7- AVhen viewed laterally, internodes of 

 this structure may exhibit either two cells, or one cell and two half cells, appearing 

 to the eye as three, the middle one twice the breadth of the others. Such stems 

 are said to be four-tubed, or bi-tri-striate. In some species which are of this cha- 

 racter small cells {secondary cells) are formed external to the four primaries^ and 

 alternating with them, as shown in PI. XYII. A. fig. 6, and B. fig. 7 ; and in 

 others, as at PL XVI. A. fig. 4, a thick epidermal layer of small cells surrounds the 

 primarif tubes, and such stems become apparently inarticulate, the articulation being 

 concealed beneath the coat of small surface cells, as at fig. 5. By turning to the 

 figures, the gradual complication of a stem of four primary tubes will be readily 

 understood ; and the same occurs where there are more primaries than four, as at 

 PI. XVI. D. fig. 5, PL XVII. C. fig. 3, which latter figure shows the commence- 

 ment of secondary tubes in a polysiphonous stem. The stem and branches of all 

 the species are formed on one or other of these models, the number of primary 

 tubes being sometimes as many as twenty-four. 



In studying the species, it is absolutely necessary to make cross sections of the 

 stem in order to be able to count the number of primary tubes in each internode ; 

 for though the exact number is not of specific importance in the polysiphonous 

 species, where we find the same species varying in different specimens from 12 to 

 20 tubes, yet it is essential to determine whether there be only four, or more than 

 four ; and, generally speaking, the species with fewest tubes are most constant to 

 their number. A little practice will enable the student to make the necessary sec- 

 tions with the help of a small-bladed knife and a botanical simple microscope, for 

 which may be substituted a watchmaker's eyeglass. A small piece of a steni or 

 branch, say a quarter of an inch long, is firmly held by the finger of the left hand 

 upon a slip of glass under the lens, while tliin slices, the thinner the better, are cut 



