78 THE CACTACEAE. 



Key to Genera continued. 



Axils of flower-scales naked. 



Tube of flower longer than limb, its scales not fimbriate; spines all 



straight 1 1 . Matucana (p. 102) 



Tube of flower not longer than limb, its scales fimhriate ; central spine 



hooked 12. Hamatocactus (p. 104) 



CC. Flowers mostly campanulate, at least not long and slender. (See 

 Leuchlenbergia and some species of Gymnocalycium) . 

 D. Areoles not arranged on ribs. 



Tubercles short, imbricated, much as in Ariocarpus 13. Strombocactus (p. 106) 



Tubercles much elongated, slender, finger-like 14. Leuchlenbergia (p. 107) 



DD. Areoles arranged on definite ribs. 



E- Scales of ovary and perianth-tube entire (except in some large 

 species of Fcrocactus) . 

 F. Axils of scales on ovary and fruit naked. 



Ribs usually continuous, rarely if ever tubercled; flowers 

 with scarcely any tube. 

 Ribs usually numerous, much compressed and thin. . . 15. Echinofossulocaclus (p. 109) 



Ribs not so numerous, usually thick 16. Ferocaclus (p. 123) 



Ribs usually broad, tubercled; flowers with a short but 

 definite tube. 

 Spines or most of them acicular; flowers purple; seeds 



muricate 17. Echinomastus (p. 147) 



Spines stout, mostly subulate; flowers white, yellow, 



or pink; seeds tuberculate 18. Gymnocalycium (p. 152) 



FF. Axils of scales on ovary hairy, woolly or setose. 

 Ovary-scales many, their axils woolly. 



Axils of scales on flower-tube neither bristly nor spiny. 

 Ribs several to many; plants very spiny. 

 Fruit permanently woolly, nearly dry, dehiscent 



by a terminal pore . . . . , 19. Echinocactus (p. 166) 



Fruit not so woolly, bursting irregularly, some- 

 what fleshy 20. Homalocephala (p. 181) 



Ribs few, broad; plants mostly spineless 21. Astrophytum (p. 182) 



Axils of scales on flower-tube usually bristly or spiny 

 as well as woolly. 



Top of fruit spinose; seeds pitted 22. Eriosyce (p. 186) 



Top of fruit not spinose; seeds not pitted. 

 Seeds not shell-like; plants mostly large. 



Spines straight or rarely curved; seeds with 



truncate base 23. Malacocarpus (p. 187) 



Spines acicular, one of centrals strongly hooked ; 



seeds minute 24. Hickenia (p. 207) 



Seeds shell-like; plants very small 25. Frailea (p. 208) 



Ovary-scales few, their axils with tufts of short hairs. 



Fruit a berry; spines all straight 26. Mila (p. 211) 



Fruit dry, dehiscent by a basal pore; some spines hooked. 27. Sclerocaclus (p. 212) 

 EE. Scales of ovary and perianth-tube fimbriate-lacerate; plant 



small; flowers nearly rotate 28. Utahia (p. 215) 



1. DENMOZA gen. nov. 



Plant cylindric, often elongated, the numerous, parallel straight ribs slightly undulate ; spines 

 in clusters at the areoles; flowers arising from the top of the plant, zygomorphic, scarlet, with a slen- 

 der throat and very narrow limb ; tube proper very short, its mouth closed with a mass of white wool ; 

 inner surface of the elongated throat covered with stamens; filaments and style long-exserted ; ovary 

 and tube bearing numerous scales, their axils filled with silky hairs; fruit globular, dry, splitting 

 down from the top; seeds black, dull, pitted. 



Type species: Echinocactus rhodacanthus Salm-Dyck. 



The generic name is an anagram of Mendoza, the province in Argentina, where the 

 plant is native. Only one species is known. 



The peculiar mass of white wool near the base of the flower-tube on the inside is not 

 known, as far as our observation goes, in any other cactus, except in two species of Lobivia 

 of doubtful relationship, described by Dr. Spegazzini as species of Echinocactus, and in E. 

 spiniflorus, all of which are otherwise quite different from Denmoza. 



The genus here segregated was considered by Schumann as a species of Echinopsis 

 but it has also been referred to Cereus, Echinocactus, Cleistocactus , and Pilocereus. In its 



