ROSA 



ROSA 



1547 



regard to the foliage the handsomest of the hardy 

 Roses, with its dark green leathery and glossy leaves. 



Most of the species are hardy or almost hardy north, 

 as A', rugosa, setigera, Carolina, Virginiana, lucida, 

 hnmilis, canina, rnbiginoaa, spinosissima, alpina, ar- 

 I'ensis and ninUiflora. Some species, as -B. Wichurai- 

 ana, sempervirens, seHcea, microphylla, Chinensis and 

 JSglanteria, require protection north. Others, as H. 

 Banksia, hracteata, Iwvigata and gigantea, are hardy 

 only south. 



With few exceptions the Roses are of easy cultivation 

 and grow in almost any kind of soil, except in a loose 

 and very sandy one. They are readily transplanted. 

 The Wild Roses need little pruning; they should only 

 be thinned out and the weak and old wood be removed; 

 long and vigorous shoots should not be shortened, es- 

 pecially in the climbing varieties, as these shoots are 

 the most floriferous. 



All true species can be propagated by seeds. The 

 hips should be gathered as soon as ripe, the seeds 

 washed out and sown at once or stratified and sown in 

 spring. They germinate the first year, but if kept in 

 the hips during the winter and allowed to become dry, 

 they usually do not germinate until the second year. 

 Mice are very fond of the seeds. Almost all species 

 grow readily from cuttings of nearly ripened wood in 

 summer under glass. Many species, especially the 

 climbing Roses, can be propagated by hardwood cut- 

 tings taken in fall and planted in spring. Layering is 

 less often practiced, except with a few species, like M. 

 Ititea and 11. hemisphwrica, which do not grow readily 

 from cuttings. Some species, especially those of the 

 groups of Cinnamomeffi, Carolinae and Gallicse, can be 

 increased by root-cuttings; the roots are taken up in 

 fall, stored during the winter in sphagnum or sand in a 

 frost-proof room, and sown in spring in drills and 

 covered about 2 inches deep. The species of the last- 

 named groups and some others are also often increased 

 by suckers and division. Budding and grafting is less 

 often done with the W^ild Roses and should be avoided 

 for Roses in shrubberies where the individual plants 

 cannot be carefully watched; the stock usually throws 

 up suckers and outgrows the cion, often in a short time. 



and species the innumerable forms which often pass 

 gradually into each other. In no other genus, perhaps, 

 are the opinions of botanists so much at variance in 

 regard to the number of species. While some, as 

 Bentham and Hooker, estimate the number at about 30, 



2146. A 5-foliolate Rose leaf. 



Rosa is a widespread genus, easily distinguished by 

 well-marked characters from allied genera, but in the 

 limits of the genus itself the characters are exceedinglj' 

 variable and it is very difficult to group into sections 



2147. A 9-foliolate Rose leaf. 



the French botanist Gandoger actually describes from 

 Europe and western Asia alone 4,266 species. The 

 majority of botanists recognize over 100 species. The 

 Roses are almost equally distributed through the colder 

 and temperate regions of the northern hemisphere, in 

 America extending to North Mexico, in Africa to Abys- 

 sinia, and in Asia to India. They are all shrubs of 

 upright habit, or climbing or sarmentose, with usually 

 prickly stems: Ivs. stipulate, alternate, odd-pinnate, 

 with 3 to many Ifts. (Figs. 2146, 2147), rarely simple: 

 the fls. are mostly large and showy, pink, purple, white 

 or yellow, and appear usually solitary or corymbose at 

 the end of short branchlets ; petals and sepals 5, rarely 

 4; stamens numerous; pistils numerous, rarely few, in- 

 closed in an urn-shaped receptacle, which becomes fleshy 

 and berry-like at maturity, containing several or many 

 bony akenes, usually erroneously called seeds; the fr. 

 itself is called a "hip." Pig. 2148, 2149. The fls. show 

 a remarkable tendency to become double, and such 

 forms have been known and cultivated from time im- 

 memorial. These innumerable garden forms, increas- 

 ing every year, are almost exclusively of hybrid origin 

 and are therefore omitted in the botanical classification 

 of the genus. 



Many attempts have been made to subdivide the 

 genus with more or less satisfactory results ; the more 

 important are those by A. DeCandolle, Lindley, Regel 

 and Baker. Nowadays the arrangement proposed by 

 Cr^pin is considered the most natural and satisfactory 

 and has been followed in the account given below. No 

 good general monograph has been published since 

 Lindley 's Monographia Rosarum (1820), except a rather 

 short one by Regel in 1877. Of the more recent publi- 

 cations the most important are those of Cr^pin, espe- 

 cially his "PrimitisB Monographise Rosarum." In con- 

 sulting his publications one has to bear in mind that the 

 author changed his opinion somewhat respecting the 

 value of the species during his studies of the genus. 

 In his later publications he takes a broader view in 

 regard to the specific value of the Rose forms and 

 unites under one species many forms which he for- 

 merly considered as distinct species. An illustrated 

 monograph valuable for the knowledge of the older 

 garden forms and species is Thory and Redout^'s "Lea 



