ADDITIONS TO THE CACTUS-FLORA OF THE UNITED STATES, 225 
From other sources I am enabled to give the following further Additions and Corrections to my 
former publications : — 
Many ELumamillarie (Synops. Cact., p. 4) have an ovariwm exsertum. Not only do the large-flowered Lon- 
gimamme, which approach closely to Coryphantha, deviate in this respect from the assumed character of the sub- 
genus, but in a great many other species I find the same peculiarity ; so that I am inclined to restrict the ovarium 
immersum to that natural subdivision, the Lactescentes, already recognized by Zuccarini. Probably all those with 
limpid juice have an exsert ovary. 
MAMILLARIA BARBATA, Engelm. This species is easily propagated by seed, and apt to flower in the 
second year. The first flowers in spring (May) appear in the axils of the last, innermost tubercles of the last year, 
and are therefore ao central ; the later ones seem to be developed from the axils of the first tubercles of the 
same spring! Flowers 9-10 lines long, of the same diameter; tube constricted above the exsert oval ovary; 
12-13 exterior green sepals, lanceolate, cuspidate, fimbriate, 8 interior ones, reddish; longer, lance-linear, 
slightly ciliate ; 18-21 petals, rose-red, with a deeper colored streak, lance-linear, shorter and narrower than [202] 
the inner sepals, entire; stamens not half as long as petals, with oval anthers ; style much longer than sta- 
mens, with 5-6 short, greenish-yellow suberect stigmas. 
M. sicotor, Lehm., is not a Texan plant, as has been stated a in Synops. p.7. Dr. Poselger 
found it on another Rio Ceandle; between Tampico and Real del Monte, Mexico. 
M. PAPYRACANTHA, Engelm., is an Echinocactus, as stated above. 
M. RECURVISPINA, Engelm. in Cact. Mex. Bound., p. 12; Synops., p. 10. As there is already a species named 
thus by Vriese (see Walp. Rep. ie p- te I now name the Arizona species M. recurvata. M. recurva, Lehm., is a 
form of EE eee DC., fide Salm 
CEREUS VARIABILIS — thus named in Cact. Mex. Bound., p. 40, tab. 60, fig. 5-6, aa in Synops., p. 31 — is not 
Pfeiffer’s plant, figured in Abbild. 2, tab. 15, but seems to be, as regards fruit and seeds, identical with a species 
i acca by Dr. Poselger near Tampico, and decided by him to be C. princeps, Hort. Wiirzb. ex-Pfeiff. Enum., p. 108. 
Plants from the Rio Grande have repeatedly bloomed here at the late Mr. Grieve’s, and as the flower has never been 
described, I here supply the omission. Fruit and seed, obtained near Matamoras, have been described and figured in 
Mex. Bound. Cact. 1. ¢ 
ores ad apicem caulis ramorumve pauci magni albi nocturni ; ovario ovato areolis aculeolatis 25-30 stipato ; 
tubo elongato cylindrico sursum sensim ampliato areolis 16-20 vix aapaesien inferioribus aculeolatis munito ; sep- 
alis superioribus 20-25 lanceolatis patulis reflexisve ; petalis 0 pluriseriatis lineari-lanceolatis Sra ; 
staminibus superiori tubi parti gradatim adnatis; stigmatibus 1 Pete in caphation clavato-obovatum coarctatis palli 
virescentibus 
pies from July to September. Flower 7-8 inches long, 53-6 inches wide ; tube 4-5 inches long. Lower 
sepals near the well-defined upper edge of the tube reddish-green, 3-9 lines, upper ones petaloid, 9-18 lines long ; 
petals 2 inches long and about 4 lines wide. Lower part of the tube, for 2 or 2} inches, with a naked, nectariferous 
surface ; the upper part, 25-3 inches, densely beset with stamens of about equal length, so that the mass of the an 
form a deep funnel corresponding to the shape of the upper part of the tube. The outer series of stamens forms a 
regular crown, but is not separated from the inner lower ones by a naked belt, such as is found in many species ; nor 
are the filaments declined and, so to say, fasciculated. This is interesting, as it weakens the value of this arran 
ment of stamens as a generic or subgeneric character ; nevertheless, it is one of the few general characters left us [203] 
to be used in the arrangement of the very large number of species of this protean genus, to which several lately 
established genera have to be reduced. The following disposition is suggested : — 
1. Cerei flore be ed plerumque breviore ; staminibus tubo gradatim adnatis. 
2. Cerei flore swpe obliquo plerumque longiore; corona staminum exteriorum erectorum a ceteris gradatim adnatis plus 
minus declinatis discreta. ; 
Echinopsis, Zuce. 
Eue i 
Phyliocactus, Link. 
Disis k. 
Under the name Acanthocerens I comprise the species of this division with spiny fruit, but not belonging to 
Echinocereus ; it is probable that Pfeiffera, Salm, is only a diminutive form of Acanthocereus. Lepidocereus — to 
29 
