104 THE .TOUilXAL OF BOTANY 



Yar. Bloxamiamis Colem. Delete " rather short " in description 

 of panicle. 



(37 flf. R. PR.ERUPTORUM Bah. (not Boul. ?) : see Ilandhool', 

 p. 09, 1. 4 from top. 



68. R. MELAN'ODERMis Focke. Delete "short " in description of 

 panicle branches. 



(39. R. Babingtonii Bell-Salt. Remove var. i^hifUothyrsus 

 (F rider). 



(o^a. Add R. festtyus Muell. S: Wirtg.. which covers some of 

 the plants formerly put ww^tiY phyllothyrsus (Frider). Journ. Bot. 

 19U, p. 207. 



70. R. Lejeunei Wh. & N. var. ericetorum (Lefv.) is made a 

 species. Under it are placed var. cuneafus Rogers & Le^^, suhsp. 

 sci'fijlonis (P. J. Muell.) and its var. sco//cws Rogers & Lev. Journ. 

 Bot. 1906, pp. 59, 60. 



72. R. mutabilis Genev. Add var. Naldretti AVhite. Journ. 

 Bot. 1908, pp. 24, r)9. Wats. E. C. Rep. 1906-7. 



Between 75 and 76 comes R. glareosus Rogers & Marshall. 

 Journ. Bot. 1912, pp. 309, 374. Mr. Rogers's re-arrangement of 

 species there proposed is not here reproduced, because it is absent 

 from his interleaved Handbook, and apparently was not brought into 

 practical use bj him when he was in active work. 



79. R. LONGiTHYRSiGER Bab. var. botryeros Focke is raised to 

 specific rank. 



81 a. R. noRRiDTCAULTS P. J. Muell. Journ. Bot. 1906, p. 60. 



84. R. KoEiiLEKi Wh. & N. subsp. dasyphyllus Rogers is made 

 a full species. It is not endemic : Mr. Rogers has seen specimens 

 from Denmark. 



92. On R. iiiRTUS Waldst. & Kit. var. mimifiJJnnis (P. J. Muell.). 

 Mr. Rogers (1914) notes " Perhaps it had best be omitted from our 

 British list." 



93. R. ACUTTFROXS Lev. Add var. ampJifrons Ley. Journ. 

 Bot. 1902, p. 69. 



98. R. DUMETORUM Weihc. Between /^ro.r and hrltaniiicus add 

 A'ar. triangularis Ley. Journ. Bot. 1902, p. 70. 



Before fuberculatus add var. rad id if 'or mis Lev. Journ. Bot. 

 1904, p. 120. 



99. R. CORYLIFOLTUS Sm. For var. cydopliyUus substitute var. 

 conjungens Bab. 



Add form c, calcareics Rogers MS. ; distinct from both (a) sub- 

 lustris and (b) conjungens. Stem uniformly angled and more con- 

 spicuously so than in conjungens, though slenderer than in either 

 a or /;. Prk. as in b with stout base, but usually fewer and more 

 strictly contined to angles. L. 5-nate. Term. It. ovate-acuminate, 

 narrow, with close finely-pointed teeth as in a, but neither lobate 

 towards point nor cleft at base. Usually wholh^ eglandular. 



Apparently the prevailing corylifolian in the Cotteswolds, from 

 about 200' to over 7(30' alt., between Cheltenham eastward to King- 

 ham, and Aston-sub-Edge southward to Foss Cross. Fui-ther obser- 

 vations in other limestone districts may be necessary- for decision as 

 to its value as a distinct form or var. Connected bv a few inter- 

 mediates with a and &.-W. M. R., 1913. 



