MORE SIDE-LIGHTS ON THE STRUCTURE OF COMPOSITES. 39 



Ad ripas limosas fl. Cantonensis, m. Jan. 1870, stirpem mas- 

 culam primum detexit Sampson ; femineam in prov. Kwang-si, 

 juxta fl. Liang-fung m. Junio 1879, demum invenit W. Mesny. 

 (Herb, propr. n, 16446.) 



This is the plant recorded, in the fourth fasciculus of these 

 Spicilegia, as doubtfully referable to Salix ijopidifolia, Anderss.; 

 but now the fruit is discovered, it is evident this was a mistake. 

 It differs from all the specimens of S. tetraqMnna, Roxb., I have 

 seen, Chinese or Indian, in the shape of its leaves, the shorter 

 male catkins, shape of the scales, and some other particulars ; but 

 it may prove to be only a variety of that species. 



53. Abies Fortunei, Lindl. — Juxta Fu-chau, Maio 1873, legit 

 filius Alfredus ; vere 1881, W. Hancock. I believe Lindley was 

 quite right in referrmg this handsome tree to Abies, though 

 Carriere has made a special genus — lieteleeria — of it. The cone- 

 scales are not persistent, as stated by Parlatore, but they do not 

 fall so early or so readily as usual in the genus. 



54. Anactochilus. — In spongiosis umbrosissimis ditionis Ningpo- 

 ensis raro vigentem, d. 5 Octobris, 1877, invenit am. Hancock. 

 This seems close to A. Floxburghii, Lindl., and A. Reinwardtii, BL, 

 but has only two smaller flowers, and is far less robust than either, 

 as figured in Blume's splendid jDlates (Fl. Jav. iv., t. 12 & 12 b.); 

 the fimbrias are long and setaceous, as in the last-named species. 

 I have only two specimens, and cannot make a full examination 

 without destroying these. Accordmg to the collector, the blossoms 

 are pinkish white. I have seen an AiKBCtochilus in the Canton pro- 

 vince, but failed to keep it alive, or get it to flower ; and I do not 

 know if it is identical with the Ning-po plant. 



55. Polygonatum officinale, All. — In vicinibus opp. Chm-kiang, 

 prov. Kiang-su, Maio 1880, leg. Bullock. The most southerly 

 Chinese station known to me. 



56. Cheilanthes rufa, Don. — In rupe calcarea ad margines 

 cataract^e, juxta pagum Ling-kwai, secus fl. North River, prov. 

 Cantonensis, d. 30 Martii 1881, coll. rev. B. C. Henry. New to 

 the Chinese flora. The specimens are robust and well developed. 

 Mr. C. B. Clarke has remarked (Trans. Linn. Soc, 2nd ser. i. 457) 

 that this fern always grows on limestone in India. 



MORE SIDE-LIGHTS ON THE STRUCTURE OF 



COMPOSITES.* 



By Maxwell T. Masters, M.D., F.R.S. 



The Tubuhfloral division of Composites is marked, as is well 

 known, by the presence of ligulate ray-florets, which are very 

 generally female or neuter, and by tubular disc-florets, which are 

 usually hermaphrodite, at least structurally. When, as often 

 happens in cultivation, the florets of the disc lose their normally 



* See Journ. But., IbTO, pp. 33 — 36. 



