Likely to be confused with /. odorata, but the petals 

 are 4 times as long as the calyx and the pods have short 

 hairs, while in /. odorata the petals are 1% times as 

 long as the calyx and the pods glabrous. 



3 odor&ta, Linn Sweet scented or Fraqbant Can- 

 dytuft Lvs linear fls white Crete. S.B.F.G. 50. 

 Frequently confused with I pinnata. Better and more 

 fragrant in poor soil. 

 4 pinnita, Linn. Not 

 , , I advertised in America, 



'vJtil , but often sold as I. odo- 



I lata. Fls. white: in- 

 s / '/, florescence only slightly 

 y^ "/ elongated in fruit. Spain, 

 / b France, Italy. 

 / 5 Bemp6rvirens, Linn. 



\jjQg^ E\ EEGREEN C. LvS. ob- 



H"^ ff!% long, obtuse, narrowed 



t'^)> at base, glabrous: fls. 



7 white. Crete. Gng.2:145 



Z (fine habit sljetch). F.E. 



1 Id (poor). Var. pl^na, 



a double form, is cult., 



7 1 ut is less desirable. 



\ ir rdsea and var. ffiliis 



variegiltis are sold 



abroad. Var. supArba or 



^ Perfection is said to be 



one of the best forms.— 



This is the commonest, 



haidiest and most per- 



and tenderer kinds are 

 winter-killed /. semper- 

 virens is likely to spread 

 out and surround the 

 labels of other kinds. 

 1131. Iberis Gibraltarica(XK). This probably explains 

 why some of the most 



reliable dealers have sold this plant under other names, 



particularly I, Gibra,ltarica. 



6. saxAtilis, Linn. Lvs. glabrous or ciliate: fls. white. 

 S. Eu. 



Var. corlfdlia, Sims (/. corifblia. Sweet). Lvs. gla- 

 brous: fls. white. B. M. 1642, though this picture was 

 ■doubtfully referred by Baker to /. Garrexiana. 



7. Garrexi4na, AH., not Scop. Lvs. glabrous : fls. 

 white. Piedmont, Pyrenees. Referred by Index Kewen- 

 sis to I. sempervirens. Intermediate between /. sem- 

 pervirens and /. saxatilis, having the habit of the latter. 



8. Gibraltdrica, Linn. Fig. 1121. Lvs. wedge-shaped, 

 obtuse, subciliate : outer fls. pink, inner ones white. 

 Gibraltar. B.M. 124. Gn. 10:38. R.H. 1870:330. Gn. 

 24, p. 549, same as R.H. 1885, p. 446.-This is considered 

 by some as the most striking and showy of the peren- 

 nial kinds. It grows higher and more erect, with larger 

 clusters and larger fls., but is less hardy than the others. 

 This is much sought after, and the stock in the nur- 

 series is often not true to name. Var. h^brida is adver- 

 tised. 



9. umbell4ta, Linn. Lvs. lanceolate, acuminate, 

 lower ones serrate, upper ones entire : fls. in the wild 

 typically purplish, rarely white : pods acutely 2-lobed. 

 Italy, Crete, Spain. B.M. 106. -This is the common an- 

 nual Candytuft with colored fls., the colors being more 

 numerous and better fixed than in any other species. 

 American trade names are vars. carminea, cAmea, lila- 

 cina and Diinnetti (/. Dunnetti, Hort.), the last being 

 dark purple. Vars. rdsea, purpurea and Alba are adver- 

 tised abroad, also vars. nslna, ptimila and hjbrlda. Tall 

 and dwarf forms of all the colors are pnocurable. 



10. Tenoreina, DC. Lower lvs. obovate, narrowed at 

 base: upper lvs. oblong-linear: fls. purplish or whitish: 

 pods notched at apex. Naples. B.M. 2783. L.B.C. 

 18:1721. According to Baker (G.C. 1868:711), this is the 

 only perennial kind that is decidedly hairy. DeCan- 

 dolle says the lvs. are puberulous. 



11. Prtiiti, Tineo. Lvs. glabrous, obovate-spatulate, 

 entire or subdentate : fls. white: pods merely notched 

 at apex. Sicily. Not advertised here, but cult, abroad. 



IDAHO 795 



12. semp^rflorens, Linn. Lvs. wedge-shaped or spatu- 

 late, obtuse, entire, glabrous: pods scarcely notched at 

 apex. Sicily and perhaps Persia. The characters in the 

 key under D and dd distinguish this from all the other 

 species of Iberis. Once advertised by Pitcher & 

 Manda, together with var. plena, a double variety. Var. 

 foliis variegitis said to be cult, abroad. 



1. cariAcea, once advertised by Saul, is presumably a typo- 

 graphical error.—/, cordi/ulia is a f recjuent error for 1. corif olia. 

 —I.correfefdlia, Hort., is a common trade name abroad, which 

 is usually spelled 1. correefolia in American catalogues. There 

 is no genus Corra, and Correa is an Australian plant of the 

 Kutaoeffl. Specimens should therefore be compared with I. sax- 

 atilis, var. corif olia. Mottet's description, however, would place 

 this plant directly after Garrexiana in the key, being distin- 

 guished from Garrexiana by the flowers becoming purplish in- 

 stead of always remaining white. Mottet says that 1. correas- 

 folia, Hort., is a hybrid, witli spatulate, entire, obtuse lvs. This 

 Question could be quickly settled if seedsmen would keep dried 

 specimens of their plants.—/. Ihlrica, at Jolin Sard's catalogue, 

 1893, is not in Index Kewensis.— /. Ulaclna of careless trade 

 catalogues is presumably a lilac-fld. variety of I.umbellata.— 

 /. nana hObrida, Hort., is not I. nana, All., a distinr-t botanical 

 species, but a trade name of mixed dwarf varieties of some com- 

 mon annual kind, presumably I. umbeUata. w M 



ICE PLANT is Mesembryanlhemum crijstallinum. 



IDAHO, HORTICULTURE IN. Fig. 1122. The state 

 of Idaho lies entirely west of the Rocky Mountain range, 

 whose summit line forms the northeastern boundary. 

 All drainage and waterways of the state finally reach 

 the Columbia river by many directions and extensions 

 of numerous rivers ami creeks, excepting for a small 

 area in the Extreme sunthrastrrn i.ortion of the state, 

 which drains to the (in .M S:,lt L.ik. , in Utah. Generally 

 the state is very mounliiiiH.u^. hut. a roiisiderable area of 

 the southern portion c.Tistitul.-i the high table-lands 

 lying on both sides of the Snake river. Most of the 

 state lies above an altitude of 2,000 feet. At and near 

 Lewiston, in the valleys of the Snake and Clearwater 

 rivers, the altitude drops suddenly to 647 feet and up- 

 wards. The numerous mountain chains and peaks which 

 cover this vast Rocky Mountain slope, direct the streams 



in endless ways to their outlets into the large rivers. 

 Thus it can be understood that climatic influences are 

 extremely variable. Altitude does not altogether deter- 

 mine the character of the climate in the valleys. The 

 prevailing currents of air in a given locality are often 

 influenced and directed by the direction of the mountain 

 ranges and the proximity of snow-clad peaks. Greater 



