UC-NRLF 
 
 SB 
 
 H- 1 3 
 
 73 22fl 
 
 
Price to Fellows, 2s. 6d. ; to Non-Fellows, 5s. 
 
 NOV 2 
 
 Dortfcultural Society 
 
 BULBOUS IBISES. 
 
 BY 
 
 PROF. MICHAEL FOSTER, Sficli 
 
 ,^^ 
 
 Published at the Society's Offices, 11 7 Victoria Street, Westminster, S.W. 
 
 printed for tbe IRogal horticultural Society 
 
 BT 
 
 TISWOODE & CO., NEW- STREET SQUARE, LONI 
 
 All applications for Advertisement* must be m^de to The Advertisement Manager, 
 Royal Horticultural Society's Publications, 168 Fleet Street, London, E.C. 
 
ADVERTISEMENTS. 
 
 KENT, the Garden of England 
 
 UNEQUALLED FRUIT TREES. 
 
 Gentlemen planting should inspect our stoc 
 of over half a million Trees during the Autumn. 
 
 Liberal Terms, 
 
 Free Carriage. 
 
 Discounts* 
 
 E, 
 
 CULTIVATE 800 
 
 TRUE TO 
 
 o 
 fe 
 
 The Champion Exhibitors of Hardy Fruits, 
 
 STRAWBERRIES 
 
 NEW DESCRIPTIVE CATALOGUE FREE IN JULY. 
 
 ALL GOOD NOVELTIES, AND EVERY "WORTHY OLD VARIED 
 FRUIT TREES TRAINED FOR ALL PURPOSES. 
 
 Pot Orchard House Trees, Figs, Vines, &c. 
 
 MERCHANT NURSERYMEN ; ROSARIANS, POMOLOGISTS 
 FLORISTS AND SEEDSMEN, 
 
 CATALOGUES FOR ALL DEPARTMENTS, 
 
1Ro?al Ibotttcultural Socfet?. 
 
 BULBOUS IBISES. 
 
 BY 
 
 PROF. MICHAEL FOSTEK, SEC.E.S. F.E.H.S. &c. 
 
 PUBLISHED AT THE SOCIETY'S OFFICES, 117 VICTORIA STREET, WESTMINSTER, S.W. 
 
 printed for tbe "Regal Iborttcultural Society 
 
 BY 
 
 SPOTTISWOODE & CO. NEW- STREET SQUARE, LONDON. 
 

BULBOUS IBISES. 
 
 By Professor M. FOSTEE, F.B.S. 
 [A Lecture delivered May 3, 1892.] 
 
 [The following contains the substance of the lecture as given, 
 though I have somewhat expanded and variously modified what 
 I actually said ; and I have added a detailed description of the 
 several species. 
 
 I am much indebted to the editors of the Gardeners' 
 Chronicle and of the Garden for the loan of woodcuts ; the 
 source is indicated in each figure. Those figures which bear the 
 name " Caparn " were most kindly drawn for me by my friend 
 Mr. J. W. Caparn, of Oundle. The rest were prepared for me 
 by Mr. Wilson, of Cambridge. Unless otherwise stated, the 
 figures are of natural size.] 
 
 I PROPOSE to treat my subject as much as possible from a 
 gardener's point of view, and shall therefore not take up more 
 time than is necessary with botanical details. To start with, 
 I use the term " bulbous " in the gardener's, not in the botanist's 
 meaning. We gardeners are regarded, and indeed justly re- 
 garded, by the botanists as being very loose in our use of the 
 word " bulb "; we often include as bulbs what ought to be called 
 " corms," " tubers," or the like. I do not propose to discuss 
 to-day which Irises form " true " bulbs, and shall use the term 
 " bulb " in the loose gardening sense. If a gardening definition 
 of a bulb be wanted, we may perhaps say that " it is a specially fed 
 bud which separates of its own accord from the mother stock 
 in order to live an independent existence." 
 
 Iris Sisyrinchium. 
 
 Let me begin by calling your attention to a little bulbous 
 Iris which is perhaps the most widespread, geographically 
 speaking, of the whole genus, and at the same time probably 
 one of the oldest Irises in existence, retaining the archaic 
 features of the stock from which many other Irises have 
 
 A2 
 
 364294 
 
descended. This is Iris Sisyrinchium (figs. 1 and 2), which may 
 be traced from Spain, Portugal, and Morocco on the extreme west 
 of Europe, through Algiers, Sicily, Corsica, South Italy, Greece, 
 Egypt, Palestine, Asia Minor, Persia, and Beloochistan to 
 Afghanistan and the Punjaub in the east. 
 
 The bulb, small and globose that is to say, a flattened sphere 
 may be readily recognised by its shaggy coats, several of which 
 
 may be peeled off from 
 an old bulb. Hence the 
 name ; for the Greek word 
 sisyra (fftarvpa) means a 
 shaggy goat skin. From 
 between two very narrow 
 channelled curving leaves 
 springs a slender stem, 
 half a foot or a foot, or 
 even more in height ; 
 this bears at its summit, 
 and sometimes on late- 
 rals, a tuft of small, 
 purple, but variously 
 marked, often fragrant 
 flowers. Each flower is 
 very fugitive ; it does not 
 last even a day. Open- 
 ing in the late forenoon, 
 it closes and withers in 
 the afternoon, so that the 
 owner who is away all 
 day never sees it ; but, 
 to atone for Aiis, many 
 blooms open in succes- 
 sion. It varies much in stature, and in the colour and marking 
 of the flowers. One of the most distinct varieties is the Iris 
 maricoides of Regel, which I feel unable to sever from the rest 
 as a true species. Another is the I. monophylla of Boissier and 
 Heldreich, which has one leaf only, frequently one flower only, 
 and a stem a few inches high. Considering, however, the wide 
 geographical distribution of the species, it is to be wondered that 
 it does not vary more. 
 
 FIG. 1. IRIS SISYRINCHIUM (reduced 
 one-fourth). 
 
3 
 
 In one point of intimate structure namely, the coalescence 
 of the filaments of the stamens in their lower parts with each 
 other and with the styles it shows its affinity to some of the 
 South African Irids, for instance, to the genus Moraea; and, 
 indeed, by some authors the plant is spoken of as a Moraea, and 
 not as an Iris. But I will not dwell on this, except to say that 
 the amount of coalescence is variable. 
 
 As a rule, this Iris does not do well in this country. In all 
 its natural habitats it is 
 sent to rest after flower- 
 ing, being dried and 
 baked by the heat of a 
 dry, hot, rainless summer; 
 this it does not naturally 
 meet with in our land. 
 And, though it is a pretty 
 little thing, it hardly re- 
 pays the efforts necessary 
 to give it artificially such 
 a summer. Whenever I 
 have been able to get in- 
 formation as to the soil 
 in which it lives at home, 
 that soil has been, in 
 nearly all cases, not 
 " peaty sand," but a stiff 
 clay, baked to a hard 
 brick in summer. And 
 though I cannot pretend 
 to have had much success 
 in cultivating it, it has 
 seemed to me happiest 
 when grown in the stiffest 
 soil at my command. Pro- 
 bably in most places in this country the best plan would be 
 to take the bulbs up after flowering, and replant in autumn. In 
 any case it needs the hottest, sunniest spot. 
 
 The Eeticulata Group. 
 
 I have suggested that I. Sisyrinchium is a very ancient 
 primitive bulbous Iris, a bulbous Iris which still retains many 
 
 FIG. 2. IBIS SISYRINCHIUM. 
 
of the characters of the bulbous Irises of long ago. And 
 we may fairly suppose that from it has descended the more 
 specialised form now so well known as I. reticulata (fig. 3). 
 This receives its name from the netted nature of the coats of the 
 bulb * (figs. 4 and 5). I. Sisyrinchium has also netted coats, 
 but the coats of I. reticulata are fewer and thinner than those 
 of I. Sisyrinchium, never forming a shaggy envelope, and the 
 pattern of the network is different. The form of the reticulata 
 bulb is, moreover, an oval, sometimes a long oval, not a flattened 
 
 FIG. 4. BULB OF IRIS KETICULATA 
 (Caparn). 
 
 FlG. 3. IBIS BETICULATA. FlG.'5. POBTION OF THE OuTEB 
 
 (From the Garden.) COAT OF A BETICULATA BULB, 
 
 magnified five times. 
 
 sphere as in Sisyrinchium, so that there is no difficulty in dis- 
 tinguishing the bulbs. 
 
 I have used the phrase Iris reticulata as if there were a 
 single species only ; but we now know several allied Irises, arid 
 the number will probably be still enlarged, forming together 
 what may be called " the reticulata group." 
 
 * The netted character is more marked in the outer older coats than in 
 thei nner newer ones, so that if the outer coats be removed it is not so easy 
 to recognise that the coats are netted. This is more particularly so with 
 some of the species of the group e.g. I. Danfordice and L Kolpakowskiana, 
 in which the inner coats show hardly any netting at all. 
 
What is generally considered as the typical I. reticulata 
 is a familiar and beloved plant. Nearly all gardeners know 
 its beautiful deep violet fragrant flowers, with the long narrow 
 fall* stretched out nearly on the level, and bearing just in 
 front of the stigma a bright golden or orange patch, which, 
 forming a charming contrast to the deep violet of the rest of 
 the blade, serves as a "signal" for the insects, showing them 
 the mouth of the path which leads to the nectar at the stamen's 
 base. Also well known is the less handsome form known as the 
 variety Krelagei, with its broader segments and more purple, 
 not unfrequently dull-coloured flowers, which, moreover, are 
 rarely fragrant. The former is called the type, because it is 
 the one which was first described, namely, by Marschal von 
 Bieberstein, early in this century (having been discovered, Sweet 
 says, by Adams, who sent the plant to Bieberstein), whereas the 
 variety Krelagei was not described (and then by Regel) until long 
 afterwards. But in the native country of the two, the south- 
 eastern regions of the Caucasus, the variety Krelagei is not only 
 much more variable, but also far more abundant than the type. 
 The variety Krelagei is abundant in the neighbourhood of Tiflis, 
 and all the specimens of imported bulbs during the last few 
 years which have come into my hands have been of this variety ; 
 so far I have not come across any specimen of the type brought 
 direct from its native habitat. Moreover, seedlings of the type 
 turn out in many cases to have all the characters of the variety 
 
 * I use the familiar term "fall " instead of " outer perianth segment " 
 or "sepal," and in like manner " standard " instead of "inner perianth 
 segment " or " petal." Each consists of a terminal " blade," usually ex- 
 panded, and attached to the rest of the flower by a "claw" or "haft," 
 which is usually narrower than the blade, and in the case of the fall 
 frequently marked off from it by a constriction. The three parts forming 
 the centre of the flower are sometimes called " petaloid stigmas " ; each 
 consists of a body overhanging the stamen, united at the base with its 
 fellows into a column, and ending above in two triangular, quadrate, or other- 
 wise shaped " crests " of variable size. Immediately below the crests is a 
 horizontal ledge of variable size and form projecting towards the blade of 
 the fall, and so arranged that an insect brushes against the upper surface as 
 it crawls into the sort of tunnel formed by the body of the style above 
 and the claw of the fall below. It is this upper surface, and this only, which 
 is the true stigma, and on which the insect visiting the flower deposits the 
 pollen which it has gathered from another flower; all the rest is simply 
 " style." We may imagine the body or stem of the style to divide at its upper 
 end into three leaves or branches, one of which becomes the " stigma " proper, 
 while the other two are modified into the often very conspicuous "crests." 
 Since the ovary is composed of three united carpels, each of which has a 
 right to a style, I prefer to speak of three styles united at their bases, rather 
 than, as Mr. Baker does, of one style dividing into three " style branches." 
 
6 
 
 Krelagei, even when they are raised under conditions which seem 
 to exclude the possibility of the parent having been crossed with 
 Krelagei. For these reasons I am inclined to think that what 
 we call the type is a special variety or sport, which Bieberstein 
 happened to come across, and what we call the variety Krelagei 
 ought to be considered as the type. 
 
 Be that as it may, I. reticulata, both in its typical form and 
 in the variety Krelagei, is characterised not only by the netted 
 coats of the bulb, but also by the peculiar form of the leaf, which 
 is four- sided, square in section, and armed at the tip with a 
 horny point. The only Iris outside the reticulata group possess- 
 ing such a leaf is the peculiar Iris tuberosa of Southern Europe, 
 of which I shall speak later on. The flower is sessile, but is 
 thrown up above the as yet short leaves by means of the long 
 tube, which is longer in the type than in the variety Krelagei. 
 
 The variety Krelagei differs from the type not only in colour 
 and in form, but in the distinctness of the lateral veins on the 
 claw of the fall ; these, in the type, are lost in the general ground- 
 colour. In the type the ripe pod is long and narrow, thrown up 
 above the ground by a somewhat long pedicel ; in Krelagei the 
 pod is short and broad, and lies half buried in the ground. 
 
 The variety Krelagei varies very much in colour, being some- 
 times of a dull plum colour, sometimes of a rich red-purple, 
 sometimes almost black ; and it varies also much in size. From 
 the type the late Mr. Nelson raised two varieties having all the 
 general features of the type, except that in one the flower was 
 larger than in imported specimens, and in the other the colour 
 was not violet, but a peculiar light blue, varying, however, a good 
 deal in tint. Mr. Max Leichtlin has also raised some seedlings of 
 various shades of blue. All these are very beautiful, g,nd one of 
 Mr. Max Leichtlin's forms is an especially handsome flower. 
 These more or less light blue varieties of the typical form are 
 sometimes spoken of as the variety ccerulea. 
 
 A few years ago there was introduced, under the name of 
 I. reticulata cyanea, a small dwarf variety which differs from 
 both the type and Krelagei, not only in being of a very striking 
 blue, the blue known as cyanic, verging towards indigo, but 
 also in form, size, stature, and in some other minor features. 
 Three or four years ago I received, through the kindness of 
 Dr. Baynolds, of Van, some specimens gathered near Van, in 
 
Armenia, which resembled the above in form and other features, 
 but which were purple in colour; and I have quite recently 
 received from Mr. Allen, of Shepton Mallet, a plant also like the 
 above in form and other features, but of a very pale blue colour. 
 We obviously have to deal here with a distinct third variety of 
 reticulata, which may or may not possess the distinct blue 
 colour justifying the name cyanea. 
 
 A few years ago I received through the kindness of Mrs. 
 Barnum, of Kharput, a fourth variety, which I described in 
 the Gardeners' Chronicle as var. sophenensis. This is charac- 
 terised by the narrowness of the segments and the metallic sheen 
 of the coloration, as well as by the fact that the flowers expand 
 as soon as, or even before, the leaves pierce the soil. This variety 
 seems to occur in several shades of colour, from a red-purple 
 to a lightish blue. It is fairly abundant in Central Asia Minor. 
 
 Quite recently Mr. Max Leichtlin has introduced, also from 
 Central Asia Minor, under the name of var. purpurea, a fifth 
 variety, which resembles Krelagei in colour, but in many of its 
 features comes near to sophenensis. 
 
 And there is yet a sixth variety. But before I speak of this 
 I must turn to a member of the reticulata group, found in 
 Palestine, which has been described as a distinct species, under 
 the name of I. Histrio (fig. 6). This, however, does not seem to 
 me to differ more widely from the varieties of 7. reticulata, of 
 which I have just spoken, than do some of these from each other, 
 and I therefore cannot help regarding it as in reality a variety 
 of reticulata. Indeed it differs from reticulata chiefly by its 
 colour, which is peculiar. The fall in its central parts is of 
 creamy white, dotted over with blotches of a bright blue ; these 
 blotches fuse together at the edge and tip of the fall into a uni- 
 form ground colour of blue ; the blue, however, is not a pure 
 blue, having a slight admixture of red. There are certain 
 differences in form between it and reticulata, but these are not 
 marked ; more distinctive, perhaps, is the habit possessed by the 
 plant, that, unlike the other varieties of reticulata, the leaves 
 acquire very considerable length before the flower expands. 
 
 I. Histrio is found in several places in Palestine, which country 
 is also given as one of the habitats of the typical reticulata ; 
 but I have never as yet come across bulbs of the latter, about 
 which there could be no doubt that they came from Palestine, and 
 
its supposed occurrence in that country is probably based merely 
 on herbarium specimens, which in this matter are not wholly to 
 be trusted. On the contrary, I would rather say that typical 
 reticulata represents the most eastern and Histrio the most 
 western development of the species. This view being accepted, 
 
 FIG 6. IRIS HISTEIO. (From the Gardeners' Chronicle.) 
 
 it is interesting to note that Western Asia Minor furnishes us with 
 a form intermediate between the more ordinary reticulata and 
 Histrio. This was brought to my notice some years ago through 
 
9 
 
 the kindness of Miss Wright, of the American Mission in Amasia, 
 and since then Mr. Max Leichtlin has obtained a large supply 
 of it. This, the sixth variety of reticulata referred to above, in 
 many specimens very strikingly resembles Histrio in its colour, 
 being marked in the same way with bright blue spots and blotches ; 
 hence I propose to call it I. reticulata var. histrioides (fig. 7). 
 
 FIG. 7. IRIS RETICULATA var. HISTRIOIDES. 
 
 It differs, however, from I. Histrio in exact tint of colour, in form, 
 and in minor features, but especially in its habits. The flowers ex- 
 pand, like those of sophenensis, while the leaves are as yet hardly 
 
10 
 
 above ground ; it flowers distinctly later than Histrio, though 
 
 often earlier than the 
 ordinary reticulata, 
 and, unlike Histrio, 
 is perfectly hardy in 
 this country. More- 
 over, though some 
 specimens are spot- 
 ted and 
 blotched 
 very like 
 Histrio, 
 the plant 
 
 as a whole is very variable 
 in colour, and some speci- 
 mens are of a uniform blue, 
 marked only by deeper veins. 
 The flowers are often large, 
 larger than in any other 
 form of reticulata, the segments 
 being broad, broader even than those 
 of the variety Krelagei, and the foliage 
 is remarkably stout. It is a very 
 handsome plant, and a great addition 
 to our gardens. 
 
 So far for the actual species, 
 I. reticulata and its varieties ; but 
 we now know three other forms, 
 which, though they differ specifically 
 from I. reticulata, must be con- 
 sidered as belonging to the reticulata 
 group. 
 
 From the neighbourhood of 
 Nazareth I obtained, through the 
 kindness of Dr. Vartan of that city, 
 an Iris (fig. 8) which in dried herba- 
 rium specimens has probably been 
 confounded with I. reticulata, but 
 which when cultivated proves to be 
 
 FIG. 8. IRIS VABTANI 
 (Caparn). 
 
 very distinct. Itlias netted coats to the bulbs, and four-sided leaves, 
 
11 
 
 each armed with a horny tip, but in the form of the flower, and 
 especially in the narrow claw of the fall expanding into an oval 
 blade, it is very distinct ; moreover, it is unique (in the group) 
 in habit, since it flowers in late autumn or early winter, the 
 blooms often appearing in October. This habit, however, makes 
 it difficult to cultivate in this country ; and since the colour, a 
 slaty blue, has no great merit, the plant is of botanical rather than 
 gardening interest, and I need say no more about it. I have 
 called it I. Vartani. 
 
 Of great gardening value, however, is another species of the 
 group, 7. Bakeriana (figs. 9 and 10), growing in the mountains 
 
 FIG. 9. IRIS BAKERIANA 
 (reduced one-fourth). 
 
 FIG. 10. IRIS BAKERIANA (a rather 
 small specimen). 
 
 above Mardin, on the confines of Armenia and Mesopotamia. 
 This was brought to my knowledge by the Rev. Mr. Gates, of 
 Mardin, and Mr. Max Leichtlin secured a large supply of it, and 
 so introduced it into our gardens. It has bulbs with netted coats, 
 and the flower closely resembles that of reticulata, differing 
 chiefly in colour, the distinctive feature being that the deep 
 violet blade of the fall bears a central patch of creamy white or 
 yellow, sprinkled with dots or crossed with veins. Curiously 
 enough, the leaves are not four-sided, but eight-sided, or rather 
 
12 
 
 cylindrical with eight prominent spiral ridges ; hence, though 
 for gardening purposes it is a reticulata, we must botanically con- 
 sider it as a distinct species. The specimens collected for Mr. 
 Max Leichtlin vary a good deal in colour, and among them he 
 tells me a beautiful white form has made its appearance. Dr. 
 Cochrane, of Urumiah, kindly sent me some years ago bulbs 
 from that district, which have proved to be those of I. Bakeriana; 
 but the flowers differ in several respects from those found near 
 Mardin. 
 
 Lastly, still keeping within the reticulata group, I must 
 speak of a little yellow Iris growing in the South of Cilicia, im- 
 perfect specimens of which many years ago the accomplished 
 traveller and botanist, Mrs. Danford, sent to Mr. Baker. This 
 he described as I. Danfordice (fig. 11). Quite recently Dr. Born- 
 
 miiller rediscovered the same plant 
 in another region of Asia Minor, 
 namely Amasia, and Mr. Max 
 Leichtlin happily obtaining a supply, 
 distributed it under the name I. 
 Bornmulleri, given by Haussknecht. 
 Though the plants collected by 
 Bornmuller vary somewhat, and 
 Baker's description, being founded on 
 imperfect specimens, was not com- 
 plete, there can, I think, be little 
 doubt that we are dealing here with 
 the same species, and the earlier 
 name Danfordice should therefore be 
 used. The plant is in many respects 
 of great interest. The bulbs have 
 netted coats, the leaves^ are four- 
 sided, armed with a horny point, the 
 inflorescence and habit is that of I. 
 reticulata, and the flower has many of 
 
 the features of a reticulata flower. But besides the colour being, 
 except for a few dark green spots or veins on the fall and style, 
 of a rich yellow, the flower of DanfordicB stands quite apart from 
 all the forms of reticulata, in that the inner segments or standards 
 are almost abolished ; they are reduced to mere spikes, hardly 
 visible when the flower is viewed in the ordinary way. Still, on 
 
 FlG. 11. IRIS 
 
 The flower is not quite fully 
 expanded and rather small. 
 In the side sketch the mi- 
 nute standards are shown as 
 seen from above. 
 
18 
 
 account of its other characters, we must claim the plant as a mem- 
 ber of the reticulata group. The flower, though of a beautiful 
 rich colour, is small and low ; it appears, as in some other forms 
 of the group, while the leaves have as yet hardly speared. 
 
 These are the several members of the reticulata group as yet 
 known to us ; but I cannot help thinking that careful search in 
 the country lying between the coast of Palestine on the west 
 and Mid-Persia on the east will reveal to us yet other forms. 
 And even with those which we now possess a promising future 
 lies before the cultivator and the hybridiser. From seeding, and 
 especially from crossing the newer forms with the old, many 
 brilliant results may be expected ; we may be glad that the ever- 
 active Mr. Max Leichtlia is carrying on the work which the late 
 Mr. Nelson began. Among the forms we at present possess, 
 the typical reticulata, especially Nelson's large seedling, holds 
 beyond doubt the first place. Next, to my mind, comes the 
 variety histrioides in its best forms, with Bakeriana and Dan- 
 fordia, the latter so especially welcome for its yellow hue. 
 II is trio is beautiful, but too tender for common use, and hence, 
 next to the above, I should put the variety cyanea, followed by 
 some of the better forms of Krelagei, sophenensis, and purpurea. 
 As I have said, the variety Krelagei, especially as shown in 
 seedlings, is variable, and while some forms are hardly worth 
 Bowing, some of them are exceedingly beautiful, and, in my 
 eyes, rank not much below the type. 
 
 One word as to fragrance. For the development of this, in 
 any of the forms, warmth is necessary. The typical reticulata 
 is the most constantly fragrant of the group, but on a cold 
 February or March day even this, in the open, may fail to give 
 out an appreciable odour. With the other forms fragrance is 
 uncertain ; a pot of His trio in a warm greenhouse is deliciously 
 sweet, but in a cold border is almost inodorous ; Bakeriana is 
 generally fragrant, as are also many specimens of histrioides, 
 but sophenensis, purpurea, and cyanea are generally inodorous. 
 As a rule, Krelagei has no scent, but some specimens, especially 
 in warmth, are delicious ; and that, too, even though the 
 flowers may be extremely poor in colour. The occurrence of 
 fragrance is in fact extremely capricious, and no hard-and-fast 
 line can be laid down. The nature of the fragrance, the exact 
 kind of odour, differs in the different forms. 
 
14 
 
 As to time of flowering in this country, the following seems 
 to be the order, though variations occur, and the exact date will 
 of course depend upon situation and special climatic conditions. 
 The first to appear is Vartani, followed after a long interval by 
 Histrio. Then come more or less together, sometimes one and 
 sometimes another being in advance, Danfordia, Bakeriana, 
 sophenensis, histrioides, and cyanea. Krelagei and purpurea 
 are somewhat later, and the so-called type, as a rule, flowers 
 the last. 
 
 The cultivation of the reticulata group has chiefly to be 
 directed towards combating a disease, in the form of a minute 
 fungus, which attacks the bulbs when left in the ground, and 
 the presence of which, in the dry bulb, may be recognised by 
 the coats being splashed with black as if marked with ink. 
 When this disease makes its appearance the foliage prematurely 
 withers, and the bulb speedily rots away, leaving behind an im- 
 perfect husk filled with black powder. In any garden to which 
 the fungus has gained access, bulbs left in the ground soon 
 perish ; what one year is a beautiful clump full of bloom, may 
 next year be represented by one or two flowers only, or not even 
 by that. I am by principle adverse to too much meddling in the 
 garden, but, through successive heavy losses, I have been driven 
 to move all my reticulatas every year. I take the bulbs up as 
 soon as the foliage has died down, keep them for a while in dry 
 sand, and, before I replant them in fresh ground in July, go 
 carefully over them all, removing the coats which by their black 
 patches show signs of the fungus, and placing all really diseased 
 bulbs in a reserve ground by themselves. By this method I find 
 that I largely diminish the disease, though I have not as yet 
 wholly stamped it out. Sometimes one variety, sometimes 
 another, seems to succumb soonest to the enemy ; I do r|)t find 
 that any one kind permanently resists attack, but have in turn 
 lost patches of each kind. If I fancy one kind is disease-proof 
 because it stands several years, I am undeceived at last. 
 
 Beyond this, and the selection of a sunny, sheltered spot, dry, 
 or at least not too wet in winter, no special culture is required. 
 The plants will thrive in sandy peat, but they will thrive as well, 
 or even better, in stiff clay. When I have received imported 
 bulbs, the soil attached to them has generally been some kind 
 of stiff loam, and when I have sought information as to the soil 
 
15 
 
 in which wild plants have been found, the answer has usually 
 pointed to a stiff and loamy rather than to a light or sandy 
 soil. My own experience has also led me to choose for them a 
 moderately firm loam ; and, with the method of lifting annually, 
 I do not hesitate to make the ground as rich as possible. When 
 the bulbs are taken up every year I do not find those planted in 
 heavily manured ground more diseased than those planted in 
 virgin soil free from manure. 
 
 Much pleasure may be gained in the dull, dreary days of 
 December and January by growing these reticulatas in pots, and 
 flowering them in a greenhouse. For this purpose they should 
 be potted at midsummer, plunged in a cold frame, protected from 
 excessive autumnal rains and from early winter frosts, and 
 brought into a cool greenhouse just as the buds are about to 
 expand. They are somewhat difficult to manage after they have 
 flowered ; and whether they be kept in the greenhouse (where 
 their long foliage, necessarily increased in length by being " drawn 
 up," is a source of trouble), or whether they be returned to their 
 chill home of a cold frame, they never ripen bulbs as do plants 
 living in the open. Moreover, the giving them the proper quantity 
 of water, neither too much nor too little, needs very careful judg- 
 ment, and errors in this matter tend very markedly to injure the 
 bulbs. Hence it is advisable to make use each year of bulbs 
 which have been previously grown out of doors. 
 
 Several members of the reticulata group, such as sophenensis, 
 histrioides, Vartani, and others (I have not observed this in the 
 type or in Krelagei), throw out a very large number of small 
 bulbils round the base of the bulb. If these are planted 
 separately in a reserve ground, they will develop into flowering 
 bulbs in the course of two or three years. 
 
 Considering the tune of year in which they flower, the 
 reticulata Irises go to seed fairly well. In gathering seed care 
 should be taken not to overlook the seed-pods, which are often 
 more or less buried in the ground. The seed, if sown as soon as 
 ripe, will to a large extent germinate in the folio whig winter and 
 spring ; but some of it may lie dormant for two, three, or even 
 more years. Germination is more certain when the seed is sown 
 in the open than when it is sown in pots or pans, owing pro- 
 bably to adequate moisture being thus more regularly secured. 
 But the seedlings which appear in December or January from 
 
 B 
 
16 
 
 the summer sowing need protection if the winter is severe, and 
 in general the management of the seedlings is more easy in pots 
 than in the open ; a pot can, for instance, be easily preserved and 
 watched for three, four, or even a longer number of years when 
 germination is tardy, whereas it is difficult to keep a seed-patch 
 in the open, or even in a frame, clean of weeds and otherwise 
 preserved for such a length of time. Hence it is, I believe, the 
 best to sow in pots, choosing somewhat deep pots rather than 
 shallow pans, since the roots strike deep. But the young bulbs, 
 after their first year's growth, will thrive better in the open, 
 though they will benefit by protection in winter for one year 
 more. Seedlings of Vartani certainly, and ofHistrio to a large 
 degree, will need protection at all times. 
 
 Before leaving the reticulata group, I must mention a little 
 
 Iris, I. Kolpakowskyana 
 (fig. 12), an inhabitant of 
 Turkestan, named after a 
 Russian general who has 
 done much to further our 
 knowledge of the botany of 
 Central Asia. This is an 
 outlying member of the 
 group, being to a certain 
 extent intermediate between 
 I. reticulata and I. Sisy- 
 rinchium. It has netted 
 coats to the bulb, and the 
 flowers are single and sessile, 
 but in its leaves it resembles 
 I. Sisyrinchium, as indeed it 
 does also the features of the 
 flower. It is a charming 
 little plant, the falls pre- 
 senting a beautiful contrast 
 of rich red-purple and bright 
 FIG. 12. IRIS KOLPAKOWSKYANA. golden yellow ; but for some 
 
 reason or other it is most 
 
 difficult of cultivation in this country ; imported bulbs die for 
 the most part after the first year, and I much fear that it will 
 never become a garden favourite. 
 
17 
 
 I. Winkleri, also from Turkestan, described by Regel, I have 
 never seen ; though allied to the above, it appears to diverge 
 still more from the Eeticulata group, for the coats are mem- 
 branous, not netted. 
 
 The Xiphium Group. 
 
 We must now pass to another group of bulbous Irises, which 
 is as markedly western and European in geographical distribution 
 as the Eeticulata group is eastern and Asian, and which we may 
 
 FIG. 13. IRIS XIPHIUM, or SPANISH IBIS. (From the Garden.) 
 
 regard as a development in a direction different from that of the 
 Eeticulata group from a common ancestor, now represented, as 1 
 have suggested, by J. Sisyrinchium, found alike in Europe and in 
 Asia. This group I will venture to call, after its best known member, 
 the Xiphium group, in spite of the allied term " Xiphion " being 
 applied, unfortunately, I think, to the entire bulbous division of 
 Irises. Naturally enough, several members of this group have 
 
 B 2 
 
18 
 
 long been known to European gardens. The bulbs have mem- 
 branous, not netted coats ; the leaves are not four- sided, though for 
 the most part long and narrow, almost linear ; and the flowers, 
 generally two, but sometimes one only, are borne on stems of 
 some, and often of considerable, height. 
 
 Two species of this group are exceedingly well known : Iris 
 
 FIG. 14. IBIS XIPHIOIDES, or ENGLISH IRIS. (From the Garden.) 
 
 xiphium (fig. 13), the so-called Spanish Iris, whose headquarters 
 are in Spain, Portugal, and Algiers, but which stretches into 
 Southern France and Italy, and I. xiphioides (fig. 14), the so- 
 called English Iris, found in the Pyrenees. The latter came into 
 the hands of the Dutch gardeners by way of England, being 
 
19 
 
 carried from Bordeaux to Bristol, and so to Holland ; hence the 
 name. I need not dwell at length on the characters of these two 
 forms, so well known to all. 
 
 The differences between the two are many and striking. The 
 foliage in the English Iris is much broader than in the Spanish 
 Iris, and while the latter often " spears," and with me always 
 does so, in late autumn, the shoot appearing as an awl-like 
 spike, the latter does not spear until spring, and the shoot 
 on its first appearance has more the form of a nipple. The 
 parts of the flower of the Spanish Iris are narrow, rigid, formal, 
 the fall is extended more or less horizontally, and the style lies 
 close down upon the fall, so that the tunnel leading to the 
 nectar is very complete ; by reason of these features the flower 
 has a striking resemblance to that of the " spuria " group of 
 rhizomatous Irises. The parts of the English Iris are much 
 larger and broader, especially the blade of the fall, are lax, 
 with a more graceful sweep, and the edge of the fall is gene- 
 rally very wavy ; the style is often raised high above the fall, so 
 that the tunnel is a very open one. The colour of the Spanish 
 Iris, taking in all its varieties, is limited to blue, blue-purple, 
 yellow, and white, with a variable admixture of brown, a dis- 
 tinctly red-purple or red never occurring. The colour of the 
 English Iris is limited to blue, to purple of all tints from a nearly 
 pure blue to a nearly pure red, and to white, yellow being con- 
 spicuous by its absence. The bulb of the English Iris is larger 
 than that of the Spanish, and the outer coats are apt to fray out 
 into fibres, especially in their upper parts, so that the bulb becomes 
 shaggy ; the coats of the Spanish Iris are always much smoother. 
 The ripe capsule of the Spanish Iris is long and narrow, generally 
 more or less club-shaped, broader above than below, and the three 
 sides are deeply grooved ; the sides of the young ovary already 
 possess these grooves. The ripe capsule of the English Iris 
 is larger, broader, pointed above and below, more distinctly 
 triangular, with flat even sides. Lastly, while the seeds of the 
 Spanish Iris are small, and cubical or wedge-shaped, those of 
 the English Iris are larger, and oval or pyriform ; they are, 
 moreover, less numerous in the pod, and, as old Parkinson 
 observes, " rattle in the dry husk." 
 
 The wild forms of the English Iris, I. xiphioides, which have 
 come into my hands have always been of a deep rich blue, and, 
 
so far as I know, in a wild state it varies little in colour. The 
 variously tinted garden forms of which I spoke just now, the red- 
 purple and the almost red kinds, as well as the very common 
 forms in which a white ground is more or less splashed with 
 blue or blue-purple or red-purple blotches, are the outcome of 
 the repeated seeding to which this species for some two or 
 three hundred years has been subjected in cultivation. But 
 as I said, a yellow plant is unknown ; this colour has never 
 made its appearance during the many, many generations of 
 seedlings. Moreover, so far as I can ascertain, though repeated 
 seeding has produced great variety of colour, it has hardly 
 affected at all the structural characters of the plant ; the 
 various forms now cultivated, apart from size and colour, are all 
 exceedingly alike. This is interesting in connection with the 
 narrow geographical distribution of the species. Iris xiphioides, 
 indeed, may be regarded as the type of a really good species. It 
 differs from all its allies by characters so broad as to be obvious 
 to everyone ; it exhibits little or no tendency to vary, or to form 
 hybrids with other species. It at some time or other acquired 
 certain features, and those early became so rigidly fixed that it 
 speedily lost all power of adapting itself to varied circumstances, 
 and hence has proved unable to spread outside a very limited home. 
 The Spanish Iris, I. xiphium, on the other hand, has not 
 only a much wider range, spreading throughout the greater part 
 of Spain and Portugal into the African continent, and reaching 
 both into France and into Italy, but also comes very close to 
 other species ; so much so that between them and it the ques- 
 tion of specific differences is soon raised. Among the wild forms 
 two types may be recognised. In the one, the falls, which 
 are relatively narrow, spread out horizontally, the ovary pro- 
 trudes from the spathe-valves for some distance, and the prevail- 
 ing colour is blue or purple. In the other, the claws of the 
 falls, which are relatively very broad, rise up in a slanting 
 fashion, so as to form more or less of a funnel ; the flower 
 is " turbinate " ; the ovary is much less exserted, and the 
 dominant colour is yellow. The latter form is found in Portugal, 
 and hence, though the other blue form is also common in that 
 country, has been called I. lusitanica ; a variety of it, in which 
 the colour is not pure yellow, but heavily blotched with brown 
 is the I. sordida of Salisbury. 
 
21 
 
 The Spanish Iris, like the English Iris, has been largely pro- 
 pagated by seeding, and we now possess a very large number of 
 varieties of many tints of blue, blue-purple, purple, yellow, and 
 white, many of the flowers being parti-coloured, and a peculiar 
 effect being produced in some by the admixture of brown, giving 
 a bronzy hue. In all these we may recognise the two types of 
 which I just spoke variously intermingled. To those which show 
 traces of descent from the lusitanica stock such, for instance, 
 as " Sultane " the turbinate arrangement of the parts, and the 
 way in which the styles are overlapped laterally by the broad 
 claws of the falls, confer on the flower an aspect which contrasts 
 strongly with that of a flower such, for instance, as " Don 
 Carlos " having the characters of the type, the falls being nearly 
 horizontal as well as long and narrow, so that the centre of the 
 flower is much more open, much less closed up. As a rule, the 
 varieties which affect the lusitanica form also tend to be yellow, 
 and those with the more typical characters are chiefly blue ; but 
 this rule is by no means closely followed. Some very beautiful 
 varieties have the falls of a pure yellow and the standards of a 
 lovely blue. 
 
 When a number of plants [are examined, very many small 
 differences in the shape of the parts are met with, such as the 
 relative breadth and length of the fall, and the depth of the 
 constriction which separates the claw from the blade, in the 
 relative length and breadth of the standard, and in the presence 
 or absence of a notch at its apex, and in the crests of the styles, 
 which are generally broad and quadrate, but may be narrow and 
 almost triangular. The standards are sometimes widely spread 
 out, very divergent, but sometimes are connivent, almost meeting 
 in the centre ; sometimes they are very twisted, but sometimes 
 quite straight. 
 
 So far as one can judge from the old descriptions, such as 
 those of Parkinson, and from old figures, preserved in the British 
 Museum and elsewhere, several striking varieties known in old 
 times have been lost to cultivation. We possess one marked 
 variety of vigorous growth, with striking bronze flowers, com- 
 monly known as " The Thunderbolt " (fig. 15), but sometimes 
 called " sordida " : erroneously, since it has nothing to do with the 
 I. sordida of Salisbury. This, which seems to have been known 
 to Parkinson, unlike the other varieties, rarely bears seed ; and 
 
22 
 
 I have no doubt but that it is a hybrid, probably between the 
 yellow lusitamca variety and either L filifolia or I. tingitana, of 
 which I shall speak presently. This may be the Iris spectabilis 
 
 Fro. 15. IBIS XIPHIUM var. " THE THUNDERBOLT." (From the Garden.) 
 
 of Spach, which he regards as probably a hybrid between 
 I. xiphium and I. xiphioides ; but I see in " The Thunderbolt " no 
 trace of I. xiphioides, nor, indeed, have I as yet come across any 
 
28 
 
 plant showing any admixture of I. xiphium and I. xiphioides, 
 with either as seed-bearer. 
 
 Beyond the two types spoken of above, the wild forms of 
 I. xiphium do not vary much in structural features ; but Mr. 
 Maw, some years ago, sent me a plant which he had found in 
 the Sierra Nevada, having some special characters ; and two or 
 three years ago I received, by the kindness of the esteemed 
 botanist of Algiers, M. Durando, a bulbous Iris found by M. Bat- 
 tandier near the Marais de la Rassanta in Algiers. The flowers of 
 this, which are of a beautiful pure white colour, differ so markedly 
 in form from those of I. xiphium as to justify varietal, if not 
 specific, distinction. In all the ordinary forms of /. xiphium the 
 fall is fiddle- shaped in outline ; in this it is hardly more than 
 spathulate. It almost deserves the specific name of I. Battan- 
 dicri ; but, on the whole, I am inclined to speak of it as a variety 
 or sub-species. It is a very handsome plant, but, so far as my 
 experience goes, not so robust as the type. 
 
 Concerning an Iris inhabiting Algeria, and known as I. Fon- 
 tanesii, since I have not as yet been able to obtain it for cultiva- 
 tion, I will, for lack of knowledge, say nothing, except that while 
 some have apparently given this name to Algerian specimens of 
 /. xiphium, others believe it to be a form of an Iris of which I 
 will speak directly, I. tingitana ; and others, again, regard it as 
 a distinct species. For a similar lack of knowledge, I will say 
 nothing of the I. serotina of Wilkomm, since this also I have 
 never seen ; but from the description it seems hardly more than 
 a small variety of /. xiphium. 
 
 As a near ally to I. xiphium comes the I. filifolia of Boissier, 
 found in Spain, at Gibraltar, and in Morocco, and differing on 
 the one hand by possessing a perianth tube, though this feature 
 seems variable, and on the other hand chiefly by the red-purple 
 colour of the flowers, and by the great breadth of the " signal " 
 or golden patch on the blade of the fall beneath the stigma. It 
 is also less robust than I. xiphium, and has scantier foliage, 
 though a form in which the leaves are relatively large and broad 
 occurs near Tangier. It crosses readily with /. xiphium, the 
 offspring having intermediate characters. 
 
 More common than the above in Morocco is I. tingitana, 
 which has a most distinct perianth tube above the ovary, and the 
 bulbs and flowers of which are much larger than in either 
 
24 
 
 I. xiphium or I. filifolia. The segments are much larger than 
 in either of these, and the blade of the fall is more oval, more 
 lax, and with a tendency to be wavy at the edge ; in fact, the 
 flower shows a certain analogy with /. xiphioides. Indeed, 
 I. filifolia and I. tingitana may be regarded as the counterparts 
 of I. xiphium and I. xiphioides respectively ; and, geographically, 
 we have I. xiphioides as the extreme northern and I. tingitana 
 
 as the extreme southern representa- 
 tive of this section, the two being 
 separated by I. xiphium. 
 
 Differing more widely from the 
 above than any of them from each 
 other is J. Boissieri, discovered a few 
 years ago by Mr. Tait, of Oporto, in 
 the Gerez Mountains of Spain (figs. 16 
 and 17). It is not only dwarf, the short 
 stem bearing as a rule one flower 
 only, but the flower possesses a long, 
 narrow perianth tube above the ovary, 
 and the form of the segments is very 
 different from that of the other mem- 
 bers of the group, the narrow claw of 
 the fall suddenly expanding into a 
 broad blade, and the standard being 
 broad and short. A singular feature 
 of the flower is that the yellow 
 " signal " of the fall bears a number 
 of short hairs, almost forming a 
 " beard." In many respects it draws 
 near to /. Sisyrinchium, and we may 
 perhaps regard it as a remnant of an older form of Irfe which 
 was once prevalent in Spain, but which has been pushed out by 
 the newer L xiphium. The rich red-purple of its flowers makes 
 this Iris a welcome addition to our gardens. 
 
 We may place in this group too the beautiful J. juncea (fig. 18) 
 with its handsome, fragrant yellow flowers, though it differs widely 
 from I. xiphium in the characters not only of the flower, but 
 also of the bulb. The bulbs are more globose, not so elongated 
 as those of J. xiphium, and, as they grow old, become covered 
 with a nest of stiff brown coats. The flower, which possesses a 
 
 FIG. 10. IBIS BOISSIERI 
 (reduced one-fourth). 
 
25 
 
 very long, narrow perianth tube, is more graceful in form than 
 that of I. xiphium, less formal and rigid, having a more pleasing 
 sweep of outline ; and the rich golden colour, together with a 
 distinct fragrance, which is absent or extremely rare in the other 
 members of the group, renders it a most delightful plant. It 
 has, moreover, a wider geographical distribution than have its 
 fellows ; it is not only 
 found in Southern 
 Spain and in Northern 
 Africa (Algiers and 
 Morocco), but stretches 
 away through Sicily 
 to Italy, occurring in 
 the Riviera. A lemon- 
 coloured variety from 
 Africa is in cultivation, 
 under the name var. 
 numidica, but other- 
 wise this species varies 
 very little. 
 
 Turning now to the 
 
 cultivation of this Xiphium group of 
 Irises, the first thing to note is the con- 
 trast between the needs of I. xiphium, 
 the Spanish Iris, and I. xiphioides, the 
 English Iris. Both profit, as regards 
 vigour of bloom, by good exposure to 
 sunshine ; but while the Spanish Iris 
 delights in a dry spot, the English Iris 
 insists on an adequate supply of mois- 
 ture at its roots in summer. Hence, 
 as a rule, where the one thrives the 
 other fails. In my own dry garden, for FJQ 1? _ IRIS BoisgIERI 
 instance, I find no difficulty at all with 
 
 the Spanish Iris ; the bulbs multiply rapidly and bloom freely, 
 and seedlings grow apace. Provided that the plants are 
 not encroached upon by the roots of too vigorous neighbouring 
 perennials, or smothered by too luxurious annuals (for these, 
 and indeed all bulbous Irises, make but a poor fight in the 
 struggle for existence), they will flourish in the same spot for 
 
several years at least in succession. The English Iris, on the 
 other hand, I can only keep alive by careful effort ; as the soil 
 becomes dry in spring and early summer, the foliage becomes 
 pale and lags in growth, yielding only few and feeble flowers. 
 They need far more moisture than they can get by any watering 
 which I can give them. The different circumstances under 
 which the two forms naturally thrive is indicated by their 
 manner of growth. The Spanish Iris begins to shoot in late 
 autumn, and the foliage has reached some height before winter 
 sets in ; from which we may perhaps infer that in its native 
 
 FIG. 18. IRIS JUNCEA. 
 
 home it has little to fear from the soddening effects of a wet 
 winter. The English Iris, on the other hand, does not spear 
 until winter is practically over ; in its native wet habitat it has 
 learnt to avoid winter growth. Hence, as a rule, in every garden 
 it is only either the Spanish Iris or the English Iris which will 
 flourish without special care ; and one of the golden rules of 
 gardening, " Grow the plants for which your surroundings are 
 fitted," may be applied very forcibly here. If he whose garden 
 is adapted to the English Iris wishes to grow the Spanish one, 
 
27 
 
 let him take the bulbs up yearly, planting them somewhat late, 
 choosing each year, as far as may be, a sunny, dry spot where 
 the soil is a moderately light but not too sandy loam. He, on 
 the other hand, who fights against fate for the English Iris, 
 should choose his dampest but unshaded situation, giving the 
 preference as regards soil to a black vegetable mould rich in 
 humus, and supply artificially the moisture which may be lack- 
 ing to the plants while they are making foliage and preparing to 
 bloom. 
 
 The other members of the group seem always to require 
 special care. 7. filifolia and Ljuncea show more pressing needs 
 than I. xiphium ; they not only need a dry, hot spot, but they 
 must be kept dry in the summer for some time after flowering. 
 The easiest way to effect this is to lift the bulbs annually, re- 
 planting somewhat late, and they bear this shifting without 
 harm. If they are to be left in the ground, the place chosen for 
 them should be a hot, dry spot on the top of a rockery, with no 
 more soil than can be well drained in winter and baked dry 
 in summer ; but under such circumstances they are apt to be 
 starved. As far as my experience goes, they are more vigorous 
 in a moderately stiff loam than in a more distinctly sandy soil. 
 
 The cultivation of I. tingitana is peculiarly difficult in this 
 country, at least in most districts. The plants start growth 
 early, and their relatively broad, ample foliage is terribly 
 punished by winter storms. Moreover, they need genial mois- 
 ture and more decided warmth in early spring, just as they are 
 preparing to flower (for they should bloom in April or early in 
 May, long before the Spanish Iris), than they obtain in most 
 parts of this country. The plants are hardy enough, in the 
 sense that they can, unprotected, stand without injury even 
 our severest frosts ; not winter, but cold, cutting spring is their 
 enemy ; they live, but they refuse to bloom. I sent to a friend 
 on the Riviera some bulbs which had been growing and in- 
 creasing with me for several years, and yet without yielding a 
 single bloom ; even in the first spring of their sojourn in that 
 more genial land they bloomed profusely. I very much doubt 
 whether any artificial care can ever in this country supply what 
 this plant needs to bring out the magnificent blooms which it 
 ought to give. 
 
 Of the cultivation of I. Boissieri I cannot say much, except 
 
28 
 
 that the want of success which I have had with it leads me to 
 infer that it, like the English Iris, needs more moisture in spring 
 than my circumstances will admit. 
 
 Most of the Xiphium group go to seed freely, and the seed, 
 as a rule, germinates readily ; by far the greater part of the seed 
 sown in the summer as soon as ripe sprouts in the following 
 spring. The seed of I. xiphium starts very early, and when 
 the seed is sown in the open the young seedlings are apt to be 
 damaged, thrust out of the ground, and otherwise injured, or even 
 killed, by late frost ; hence, to secure a full crop of plants from a 
 batch of seeds, some protection is advisable. But when seed is 
 plentiful a little loss is of no great moment, and my practice 
 now is to sow in the open in a prepared bed, and to let the 
 seedlings remain until the bulbs are for the most part ready for 
 flowering. The seed of /. xiphioides is much later in germinating, 
 does not start until spring is fairly advanced, and hence does not 
 need any protection at all ; care, however, should be taken that 
 the young seedlings do not suffer from drought. 
 
 Iris tuberosa. 
 
 I may, perhaps, here say a few words about a charming little 
 plant, which is not a bulbous Iris in the strict sense of the word, 
 and which, indeed, by some authors is not considered an Iris at 
 all, and which yet has certain affinities with the bulbous Irises of 
 which I have been speaking. I mean the plant with lovely 
 black and green flowers, known in some of its native homes as 
 " The Widow " (la vedova), and called by the majority of authors 
 Iris tuberosa (fig. 19), but by others Hermodactylus tuberosus. It 
 was separated by Salisbury as a distinct genus, with the name just 
 given, because the ovary is not, as in Iris, divided completely into 
 three chambers by three septa or partitions meeting in the 
 middle along the whole length of the organ ; the partitions are 
 imperfect, not meeting in the upper part of the ovary, which 
 thus consists of a single chamber, partly divided by the projecting 
 partitions. Otherwise all the characters of the plant are those 
 of an Iris ; and, since the lack of complete fusion of the partitions 
 of the ovary may occur accidentally in many specimens of 
 Iris, it seems unreasonable to lay such great stress 011 this 
 feature. I shall, therefore, continue to consider it as an Iris. 
 But, as I said, it is not strictly a bulbous Iris ; if you dig up a 
 
29 
 
 plant when the foliage dies down you will find, not a bulb, but 
 an irregular brown tuber, like 
 a small, hard, deformed po- 
 tato, the mass being often 
 made up of two, three, or 
 more parts joined together 
 like the fingers of a hand, or 
 perhaps more like a star-fish. 
 The change, however, from a 
 regularly formed to an irre- 
 gular tuber is not a great 
 one ; and, indeed, if you sow 
 the seed of /. tuberosa, you 
 will find that the product of 
 the first year's, and indeed of 
 the second year's growth, is a 
 small rounded nodule which 
 you would at once say is a 
 bulb; this Iris is a bulb (in 
 the loose sense of the word) 
 when it is a baby, and be- 
 comes a tuber as it grows old. 
 We may probably infer that, 
 though we must now speak of 
 it as a tuberous Iris, it has 
 descended from ancestors 
 which were undoubtedly what 
 we should call bulbous. 
 
 The plant has one very 
 striking feature : the leaf is 
 four-sided, with a horny point, 
 like that of I. reticulata ; in- 
 deed the differences between 
 the leaves of the two plants 
 are relatively small, and a 
 casual observer might easily 
 confound the two. The 
 flower, again, in another fea- 
 ture draws near to a mem- 
 ber of the Eeticulata group, 
 
 FIG. 19. IRIS*TUBEBOSA. (From the 
 Gardeners' Chronicle.) 
 
 namely, I. Danjordia\ the inner segments or standards are 
 
30 
 
 reduced to mere bristles, so that at first sight they seem to 
 be absent. On the other hand, the plant betrays its affinities to 
 I. Sisyrinchium, in the filaments of the anthers being in part of 
 their course united together. We may place side by side with 
 these structural features the geographical distribution of the 
 species. While the Reticulata group, as we have seen, is confined 
 to the east, and the Xiphium group to the west, Iris tuber osa 
 stretches from almost the extreme west a long way towards the 
 east. Beginning at the west in Southern France, we may trace 
 it through the Riviera, Corsica, Sicily, Middle and Southern 
 Italy, past Dalmatia to Greece and the Grecian Islands, and even 
 to Turkey. So far as I know, however, it is absent from Asia 
 Minor. In width of distribution it is second only to I. Sisy- 
 rinchium, and, like that, is probably a somewhat ancient Iris. 
 
 The flowers, which are probably known to most, are singular 
 in colour. The ground colour of the fall is an olive-green, 
 which on the blade becomes a dark, almost black, velvety 
 purple ; this combination, with an occasional admixture of 
 yellowish or of bright green streaks, is to my mind, as to that of 
 many others, especially charming ; it has a beauty all its own. 
 The flower is single, borne on a stalk of variable length ; the 
 spathe-valves (one of which is often missing) are large and 
 swollen, and the relatively large swollen pod is a conspicuous 
 object when the foliage is ripening. 
 
 The flower differs in minor characters, in form and in colour, 
 in its different habitats, so much so that authors have made 
 more than one species ; but we ought probably not to consider 
 them as other than mere varieties. 
 
 As regards the cultivation of I. tuberosa, I can only say that 
 in this country it seems to need the sunniest, driest spot which 
 can be given to it ; and, so far as my experience goes, it dofts better 
 in a moderately light loam than in any other soil. Where it 
 thrives, it is perfectly hardy, in the sense that though the foliage 
 may shoot, and even acquire some height, in late autumn, the 
 severest winters leave it untouched ; but it is sorely tried by the 
 harsh spring winds and dry cold which are apt in England to 
 visit us in March and April, when it should be in flower. Yet ic 
 is exceedingly capricious. In some places it refuses to flower, 
 and, indeed, to grow. In my own garden it multiplies rapidly, 
 and, indeed, I find a difficulty in getting rid of it from any place 
 
31 
 
 in which it has been planted ; possibly my calcareous soil is 
 acceptable to it. Yet, while some years it flowers freely, in other 
 years the blooms are very scarce. On the whole, it seems to me 
 to do much better when left undisturbed in the ground year after 
 year than when it is lifted, though some of the older writers 
 recommend that it should be lifted, not every year, but every 
 three or four years. It does not lend itself readily to pot culture ; 
 at least that is my experience. 
 
 The Juno Group. 
 
 I must now pass on to another large group of bulbous Irises. 
 In the two groups which we have been considering, both the 
 Eeticulata group and the 
 Xiphium group, the bulb is 
 composed of two or three 
 thick, fleshy coats (the 
 swollen remnants of the 
 bottoms of leaves which 
 have vanished), wrapped 
 round the baby central 
 shoot, and surrounded in 
 turn by a certain number 
 of thin membranous wrap- 
 pings, varying in the dif- 
 ferent species. The bulb, 
 if examined when ripe, is 
 found to be free from all 
 roots, the old ones having 
 wholly disappeared, and the 
 new ones not yet sprouted. 
 The leaves, moreover, in all the members of the group are few, 
 often two only, and relatively long and narrow ; in many cases, 
 as we have seen, almost linear. By the possession of these 
 characters, the two groups form a single group, to which the 
 name Euxiphion has been given. 
 
 As a typical member of the other group of whicli I am about 
 to speak, let me now call your attention to an Iris, which is a very 
 old garden favourite, well known to Parkinson, Gerard, and even 
 Clusius, an Iris which has the honour of being depicted in plate 1 
 
 
 FIG. 20. IRIS PERSICA (reduced). 
 (From the Garden.) 
 
of the long, admirable series of plant portraits known now as 
 the Botanical Magazine, and which is probably familiar to all 
 as the Persian Iris, Iris persica (figs. 20 and 21). It is, like 
 the members of the Euxiphion group, a bulbous Iris ; but the 
 bulb is composed not of two or three very fleshy coats, but of 
 several less stout coats, surrounded, as in Euxiphion, by mem- 
 
 branous wrappings. If you take 
 the bulb up in summer, when the 
 foliage has died down, you will 
 find attached to the base of the 
 bulb a number of fleshy, finger- 
 like, but somewhat tapering roots, 
 each with a narrow 
 neck, easily broken at 
 its attachment to the 
 bulb. In the case of 
 purchased, stored bulbs 
 these conspicuous roots 
 have often been broken 
 off, and the bulb then 
 
 does not differ in outward appearance very 
 markedly from a Euxiphion bulb ; but when 
 the ripened bulb is taken direct from the 
 ground these fleshy roots are always present. 
 If you study the history of the plant during 
 the yearly cycle of its life, you will find that, 
 as the foliage and bloom are developed, these 
 thick roots shrink, and finally disappear; 
 when the plant is at the height of its vege- 
 tation, only their shrivelled remains are to be 
 seen. But as the leaves are withering in the 
 ripening process, new roots of the same kind 
 
 are formed - which become thick and stout > like 
 
 the new bulb which is forming while the leaves 
 of the past season fade and depart. Obviously 
 these thick finger-like roots are, like the thickened coats of the 
 bulb itself, stores of nourishment for the coming plant. In Euxi- 
 phion the plant possesses such stores only in the thick coats of 
 the bulb itself ; in Iris persica the plant can fall back upon the 
 supplementary stores afforded by these peculiar thick, fat roots. 
 
 flower). (Caparn.) 
 
33 
 
 Now these two features, the having several coats to the 
 bulb and the possession by the ripe bulb of thick store roots, 
 are common to a large number of Irises, which thus form a 
 group known as the Juno group. These two characters are, 
 moreover, accompanied by certain others. Thus the leaves are 
 usually broader and more numerous than in the Euxiphion 
 group, though this feature is somewhat variable. Again, in 
 most cases, in nearly all cases in fact, the flower has a special 
 form. The outer segment or fall, instead of having, as in most 
 Irises, the claw narrower than the blade, is broadest at the claw, 
 which is expanded sideways into two angular flanges or auricles, 
 one on each side. Further, in nearly all cases also, the inner seg- 
 ment or standard is very small, reduced often to a mere bristle, 
 and usually takes up a horizontal position, or is even turned 
 directly downwards instead of standing erect. To compensate, 
 as it were, for the smallness of the standards, the crests of the 
 styles are unusually large, and form a conspicuous part of the 
 flower. These several features, and other minor ones on which 
 I need not dwell, characterise this Juno group of bulbous Irises, 
 and may be spoken of as the " Juno characters." 
 
 Iris persica, in the typical form so long known the stock 
 which has been so long in cultivation coming, it is stated, from 
 South Persia, in the region between Murgab and Persepolis 
 (I have not come across any recent importations of this typical 
 form) is a striking and yet beautiful plant, with a deep violet, 
 almost black, patch on the lamina of the fall, forming a pleasing 
 contrast to the white or bluish-green tint of the rest of the 
 flower. In nearly all these Juno Irises the middle line of the 
 claw of the fall is raised into a ridge or crest, which fades away 
 as it passes on to the blade ; this ridge or crest has usually a 
 distinct colour of its own, and in Iris persica is a bright golden- 
 orange, setting off and set off by the deep violet of the blade 
 of the fall. 
 
 The short, hardly visible stem generally bears (in March, 
 sometimes earlier, sometimes later) one flower, which appears 
 sessile, but at times has two, or even three flowers. The leaves, 
 which have only just speared when the bloom appears, and do not 
 attain their full growth until long afterwards, are narrower than 
 in most other Junos, and the bulb when well grown is large, 
 being sometimes as big as a hen's egg. 
 
 C2 
 
34 
 
 The plants of Iris persica in cultivation are all very much alike, 
 and Miller remarked long ago that seedlings showed little or no 
 variation from the parent. Within the last few years, however, 
 we have obtained from various parts of Eastern, Central, and 
 Southern Asia Minor a number of forms differing from the typical 
 Iris persica, more particularly in colour, but also, to some extent, 
 in form. Mr. Max Leichtlin, for instance, has recently distri- 
 buted, under the name of I. persica var. purpurea, one of these 
 varieties found in several parts of Central Asia Minor, which 
 differs from the type chiefly in that nearly all the parts are of a 
 red-purple ; the blade of the fall, however, is especially dark, and 
 the orange signal on it affords the usual contrast. Neither 
 the fall nor the standard has exactly the form which prevails 
 in the type ; and did we possess only this purple variety, we 
 might perhaps grant it the dignity of a specific distinction. But 
 this is not the only variety. I have received from Mrs. Barnum, of 
 Kharput, and from the Rev. Mr. Gates, of Mardin, and Mr. Max 
 Leichtlin has received from these and other parts of Asia Minor, 
 plants having the general characters of Iris persica, but differing 
 not only widely in colour some being reddish, others violet, 
 others, again, steel-grey, or even sea-green but also considerably 
 in the form of the fall, more especially in the relative propor- 
 tions and details of the blade and the claw. Each of these, con- 
 trasted with the type, is so distinct that it might well be regarded 
 as a new species ; but this would entail the institution of a 
 dozen or more new species. Hence, as in other instances, we are 
 driven to consider the different forms as varieties only, so that 
 Iris persica is, after all, an extremely variable plant. A plant 
 growing in South Persia, below Shiraz, for bulbs of which I am 
 indebted to Mr. Isaacson, of Bushire, differs so much, from the 
 other varieties of I. persica that I am somewhat inclined to con- 
 sider it as a new species, but I am in doubt, and have pro- 
 visionally described it as merely var. Isaacsoni. None of these 
 varieties, to my eye, equal the singular and striking beauty of 
 the old typical form, but they are, nevertheless, nearly all of 
 them welcome additions to the garden. Most of them flower 
 earlier than the typical form. 
 
 Iris persica is often spoken of as "not hardy," and we are 
 recommended to grow it in a frame or greenhouse. If by " not 
 hardy" is meant " succumbs to severe winter frost," the term 
 
35 
 
 cannot be applied to Iris persica, since the bulbs underground 
 will stand, untouched, the severest frosts that ever visit England, 
 and, indeed, in many of its native homes it has to endure severer 
 frosts than those which visit this country. Nevertheless, it is in 
 many places difficult of cultivation, and for the following reasons : 
 it needs, when blooming and growing, genial mild warmth, and 
 when it has done growing, it needs to be ripened by heat and 
 drought. In this country, in most districts at least, it is pinched 
 with dry, cutting winds when it is young and tender, and drenched 
 with warm rains when it ought to be at rest ; hence bulbs, even 
 large and vigorous when planted, often refuse to flower beyond 
 the first year, and soon after disappear altogether. The Dutch 
 nurserymen, I understand, meet the difficulties of climate by 
 lifting the bulbs every year, and I believe that that is the best 
 course for most of us in this country also to observe. 
 This " climatic " treatment seems to be of much more im- 
 portance than the choice of soil. " Sandy peat " is, as usual, 
 recommended by many, but in its native home, in most cases in 
 which I have obtained information, it is found in loam, often of 
 a very stiff character ; and my own experience leads me to think 
 that the stronger soil yields the stronger plants. So far as I can 
 see, the Asia Minor varieties need the same treatment as the 
 typical form. 
 
 If we take the Asia Minor forms as mere varieties of the one 
 species I. persica, we may say that the species has a fairly wide 
 distribution. Stretching from South Persia westward along the 
 more southern parts of Asia Minor, it extends from the extreme 
 east of Persia to the extreme west of Asia Minor. In Armenia 
 and Kurdistan, more especially in their more northern parts, it 
 is accompanied, and eventually replaced, by another Juno Iris, 
 which stretches farther north than it does namely, into the Cau- 
 casus and which, having been first discovered in that country, is 
 called I. caucasica. 
 
 In the form which was first described, and which we must 
 therefore take as the type, I. caucasica is a dwarf plant. From 
 a tuft of four or six shiny, glossy, yellowish-green, ovate-lanceo- 
 late leaves, the margin of each of which is armed with a horny 
 ridge, rises a very short stem, often hardly visible, carrying one, 
 two, or three flowers having the Juno characters described above, 
 the whole flower being of a dull greenish-yellow. It is a plant 
 
36 
 
 " of botanical interest only," and well-nigh useless for garden 
 purposes. 
 
 The species stretches eastward along the Caucasus and the 
 North of Persia, but when we reach Turkestan we find the 
 typical form replaced by a larger, more vigorous, and really 
 handsome plant, I. caucasica major, or turkestanica. In 
 this, compared with the type, the foliage is more abundant and 
 ample, the stem more obvious, carrying often as many as five 
 or six flowers, each of which, while resembling the type in general 
 form, is larger, has the lateral expansions of the claw of the fall 
 more developed, and possesses a colour which, though somewhat 
 variable in exact hue, is on the whole a rich yellow. The yellow 
 is deepest on the blade of the fall, the effect being heightened by a 
 conspicuous orange ridge or crest, while the lateral expansions of 
 the claw are pale and usually transparent. The standards are, 
 as in Juno Irises in general, insignificant and extended horizon- 
 tally ; the crests of the styles, also yellow, are large and con- 
 spicuous. It is a handsome plant, well worthy of cultivation. 
 
 I have obtained from Kharput, in Armenia, another variety, 
 which resembles the above in its ample glossy foliage, and in 
 possessing an obvious stem, though this is covered by the decur- 
 rent bases of the leaves ; it differs in the flower, though large, 
 being more compact, with less prominent, and firmer, lateral ex- 
 pansions of the claw. In form it is exceedingly graceful, but, 
 unfortunately, lacks the golden colour of the Turkestan form, 
 possessing the greenish-yellow of the type. There also exist in 
 Asia Minor other forms, differing in various ways from the 
 type ; but none of these appear to be of value for the garden. 
 
 The name J. caucasica var. major has also been given to a 
 plant inhabiting Turkestan, which differs in many respects from 
 the plant which I have just described under that naiie. This 
 has also been distributed under the name of J. orchioides (fig. 22), 
 and since it seems to me to differ from I. caucasica by characters 
 of specific value, I propose to retain the latter name. In I. 
 caucasica turkestanica (I propose to use this term in order to 
 avoid the confusion attached to the word "major ") the stem, 
 though it exists, reaches at most a few inches in height, and is 
 almost wholly hidden by the bases of the leaves, the internodes 
 being invisible, or nearly so ; in /. orchioides the stem is often 
 two feet in height, and the nodes are separated by clearly visible 
 
37 
 
 internodes two inches, or even more, in length. In J. orchioides 
 the leaves lack the horny margin present in I. caucasica and in the 
 variety turkestanica ; they are also longer and narrower. In 
 J. orchioides the individual flowers have a distinct stalk, in 
 J. caucasica they are sessile. In I. caucasica and its varieties 
 the spathe-valves are inflated, in orchioides they are not. In 
 orchioides the fall lacks the lateral expansions of the claw so 
 
 FIG. 22. IBIS ORCHIOIDES. The side-sketch (A) shows, magnified twice, 
 the standard and its mode of attachment. 
 
 conspicuous in caucasica, and is long and narrow, spathulate, or 
 even strap-like. The whole flower is smaller and narrower in 
 orchioides, but this is compensated for by the rich golden colour 
 of all the parts, the blade of the fall being often marked with 
 black or purple spots. In both orchioides and caucasica the 
 
38 
 
 fall has a median ridge or crest of a deeper, more orange, colour 
 than its surroundings ; but this in the former is somewhat low 
 and simple, whereas in the latter it is large, conspicuous, and 
 often cut up into a fringe of short hairs, simulating the beard of 
 a rhizomatous Iris. The bulb of orchioides is very large, some- 
 times as large as a goose's egg, and the fleshy roots, so generally 
 characteristic of a Juno Iris, and well seen in I. caucasica, 
 
 are frequently ill-developed. Lastly, 
 while I. caucasica goes to seed 
 most freely, I. orchioides, so far 
 as my experience goes, yields seed 
 most scantily. 
 
 Besides this yellow form of 
 orchioides, there is also found in 
 Turkestan and Bokhara a plant 
 agreeing with it in all the features 
 of form and habit, differing only in 
 the colour of the flower, which, 
 instead of being uniformly yellow, 
 is of a delicate lavender colour, 
 except some yellow markings over 
 the ridge of the fall and its neigh- 
 bourhood. This, to my eyes, pecu- 
 liarly graceful and pleasing Iris has 
 been called I. caucasica var. cceru- 
 lea ; but, in accordance with what I 
 have stated above, I should prefer 
 to call it I. orchioides var. ccerulea. 
 Another variety, the var. oculata, 
 in which the yellow fall is simply 
 dotted with blue, I have laot seen 
 alive ; nor have I yet seen still 
 another variety, var. linifolia, in 
 which the leaves are extremely 
 narrow, linear in fact, the flowers 
 
 being yellow. Still other varieties probably occur in the 
 regions around Turkestan, for the species seems to be exceed- 
 ingly variable ; and though I have not yet come across wild 
 iorms distinctly intermediate between I. caucasica and J. or- 
 chioides (I have obtained them by artificial crossing), these may 
 
 FIG. 23. IRIS PALESTINA. 
 
39 
 
 exist, and all the forms may eventually have to be regarded as 
 varieties of one species. 
 
 Unlike the majority of Juno Irises, J. caucasica and J. or- 
 cliioides present few difficulties in the way of cultivation. Since 
 their foliage does not appear until the severe winter frosts are 
 over, they are quite hardy in the ordinary acceptation of the 
 word. Nor do they need, to the same extent, at least, as many 
 other Juno Irises, the " drying-up " in summer. Their leaves 
 do not wither (and this is especially true of I. orchioides) until 
 the hot days of summer are upon us, and these seem quite 
 adequate to ripen the bulbs. All that is necessary is to give 
 them a sunny, fairly warm spot, and a soil that is not too light. 
 As far as my experience goes, they nourish best in a rich, 
 somewhat stiff loam, and if I had to choose between clay and 
 sand, I should choose the former. 
 
 As I said a little while back, I. caucasica spreads from the 
 Caucasus westward into Asia Minor. Here, however, it occurs 
 sparsely ; and when we travel further westward and southward 
 we find it replaced by other species of limited range. 
 
 In Palestine, for instance, and the Sinai peninsula, the Juno 
 group of Irises is represented by I. Palestina (fig. 23). This, which 
 is found on Mount Hebron, and on Mount Carmel, in the valley 
 of the Jordan, and elsewhere, is a dwarf little Iris, having all 
 the characters of the Juno group, bearing one, two, or three small, 
 greenish-yellow, but variously marked fragrant flowers. It is 
 not very handsome, and, from a garden point of view, not worth 
 the trouble which its cultivation demands. With me the foliage 
 appears in late autumn, and the flowers in mid-winter. No 
 wonder that it is not hardy in this country ; I can only keep it 
 alive by growing it in a cool greenhouse. 
 
 Still southward of Asia Minor, but to the east of Palestine, 
 in the North of Mesopotamia, and the adjoining Armenian 
 hills, is found /. sindjarensis (fig. 24), for the recent intro- 
 duction of which we have to thank Mr. Max Leichtlin. This 
 presents certain analogies with I. orchioides ; the bulbs are very 
 large, and the leaves are long, lax, channelled, with their bases 
 clasping the stem, which is a foot or more in height and bears 
 two or three flowers. The general colour of the flower, which 
 exhibits the ordinary Juno features, is blue of a somewhat slaty 
 hue, broken by the yellow of the ridge of the fall and by 
 
40 
 
 greenish-blue veins and dots. It does not possess very striking 
 beauty, and yet is graceful ; it has the merit of being distinctly 
 fragrant, the odour reminding one of vanilla. 
 
 Like I. caucasica, it does not usually " spear " until the winter 
 frosts are over, flowering in March or April. So far as my 
 brief experience goes, it may be considered hardy, and appears 
 to need much the same treatment as I. caucasica. 
 
 Very closely allied to the above, differing chiefly in that it 
 bears several (eight or ten) flowers, and that these are of a 
 
 smoky yellowish colour, is the 
 I. fumosa growing in North 
 Syria, in the neighbourhood of 
 Aleppo and elsewhere. I can 
 say little about this, since I have 
 not yet flowered it, but imagine 
 that it is of no great value from 
 a gardening point of view. 
 
 As in the west /. caucasica 
 and I.persica give way in Pales- 
 tine to I. Pales tina, so in the 
 east, in South Persia, Beloochi- 
 stan, and Afghanistan, they are 
 replaced by other Juno Irises, 
 more especially by I. Stocksii, 
 which in several respects seems 
 analogous to I. Pales tina, but on 
 which, since it is both difficult 
 of cultivation and not strikingly 
 handsome, I need not dwell. Nor 
 need I detain you with the details of I. drepanophylla, growing in 
 Afghanistan, or of I. Aitchisoni, growing in the Punjauft, though 
 the yellow form (var. chrysantha) of the latter, should it prove 
 amenable to cultivation in this country, since it appears to be a 
 handsome plant, would be a useful addition to our gardens. It 
 has a botanical interest in being the last straggler of the Juno 
 group of Irises to the south-east. Probably, however, between 
 this on the east and I. sindjarensis on the west, in the zone 
 of South Persia and Beloochistan, there are other, as yet unknown, 
 Juno Irises besides the I. Stocksii which we know ; and, indeed, 
 it is possible that /. Stocksii itself, which has hitherto been 
 
 FIG. 24. IRIS SINDJARENSIS 
 (reduced one half). 
 
41 
 
 studied from dried specimens only, really represents more than 
 one species. 
 
 Let me now turn from these Juno Irises of little to one of 
 great garden value. Though I. caucasica or I. orchioides spreads 
 from Turkestan into Bokhara, there is found also in the latter 
 country a very beautiful Iris, I. Bosenbachiana (fig. 25), which is 
 a Juno Iris, but lacks some of the characters which I have stated 
 to be distinctive of the group. The fall never possesses the lateral 
 expansions or flanges on the claw which, as we have seen, are 
 so striking in nearly all other Juno Irises ; it is, in fact, almost 
 strap- shaped ; and the standard, though small and spreading 
 horizontally, or even deflexed, is relatively larger than in most 
 other Juno Irises. The bulb, too, has characters by which it may 
 be readily recognised ; the fleshy roots are numerous but very 
 short, frequently ovoid in form, not long and finger-like as in other 
 Juno Irises, so that at a little distance the bulb looks as if it 
 bore at its base a number of smaller whitish bulbs turned the 
 wrong way, pointing downwards instead of upwards. The plant 
 sends up its bloom while the leaves are exceedingly short, almost, 
 indeed, before they have appeared, so that the chief growth of 
 the foliage takes place after blooming is over ; and though the 
 one, two, or even three flowers which the bulb throws up are 
 really borne on a stem, this is so short that the flowers appear 
 wholly sessile. 
 
 These, however, are botanical features ; but the garden value 
 of the plant is due to the colour of the flower, which, in at least 
 a large number of cases, is of striking beauty. I make this 
 qualified statement because the species, though varying little in 
 form and not greatly in size, is exceedingly variable in colour. 
 If I were to adopt the practice common among " florists," and 
 give a separate name to each plant which differed in any way in 
 colour from its fellows, I could, I think, easily make a list of 
 something like a hundred named varieties. In fact, hardly any 
 two plants are ["exactly alike, and while some are extremely 
 handsome, others are poor, or even ugly. The dominant colour 
 is a combination of purple, yellow, and white ; in some the purple 
 is a red-purple passing into a rich crimson, in others the purple 
 is a blue-purple passing into a dull or dingy lavender ; and the 
 late Dr. von Regel made two varieties a red and a blue variety. 
 But the differences, as I have just said, are almost innumerable ; 
 
42 
 
 FIG. 25. IRIS ROSENBACHIANA (Caparn). 
 
43 
 
 one form, for instance, is nearly a pure yellow, with a few purple or 
 violet markings. In what is, perhaps, the handsomest form, the 
 blade of the claw is a rich deep crimson, except, on the one 
 hand, at the tip and margin, where it is of a pure solid white, 
 and, on the other hand, in the middle, where a large toothed 
 ridge of a rich golden yellow rises up ; the standards and the 
 upper surfaces of the styles are of a paler reddish-purple ; but 
 the under surfaces of the styles, and the claws of the falls which 
 they overhang, are of a golden hue, broken in the latter case by 
 crimson or purple veins. The combination of deep crimson with 
 bright gold, softened down by white and by several hues of purple 
 and of yellow, is remarkably effective ; and the only unfavourable 
 criticism which suggests itself is that the extremely rich colour- 
 ing of the flower, seen without foliage, or with very scanty foliage, 
 on the bare brown ground on a bleak day in spring, almost passes 
 over from beauty to untimely gaudiness. The same "motiv" 
 in colour may be recognised in the other forms, variations being 
 produced by the relative redness or blueness of the purple, and 
 by the predominance, or otherwise, of the yellow and white. 
 One, to my mind, very beautiful form is of a pure white, except 
 for a large patch of deep violet on the blade of the fall, and some 
 few veins. 
 
 The plant flowers in this country in early March, or even in 
 February, according to situation and climate, keeping company 
 with many of the reticulatas. It is quite hardy as regards winter 
 frost, the bulbs beneath the ground receiving no injury from the 
 severest English cold, but the ample foliage, the leaves being 
 broad and lax, is apt to be damaged by the later frosts and snow, 
 and by March winds ; hence the situation chosen for it should be 
 one which, while fully exposed to the sun, is well sheltered from 
 the wind. In general, the culture needed appears to be about 
 the same as that for I. caucasica, save that the plant is not so 
 robust and vigorous as is that species, more imperiously demands 
 to be kept fairly dry in summer, and I am inclined to think 
 prefers a rather lighter, but not too light soil. I feel, however, 
 that I have not as yet learnt all its requirements, for individuals 
 suddenly go wrong and disappear to a much greater extent than 
 is the case with I. caucasica. Still I have no reason to doubt that 
 with ordinary care, without special precautions, it can be success- 
 fully grown in most places in this country, and I feel sure that, 
 
44 
 
 when the best forms of it become known, it will be exceedingly 
 popular. 
 
 I. Bosenbachiana, as I have just said, shows some signs of 
 breaking away from the Juno group. Still more divergent is an 
 Iris found in Afghanistan, which was discovered by Dr. Aitchison 
 not far from Pendjeh of sinister notoriety, and which he has done 
 me the honour to name after me I. Fosteriana (fig. 26). This 
 
 we may include in the 
 Juno group, and yet 
 it shows many affini- 
 ties with the Xiphium 
 group. The leaves are 
 scanty and narrow, al- 
 most linear in fact. A 
 stem, a foot or even 
 more high, with clasp- 
 ing leaves, bears one, or 
 sometimes two, flowers. 
 The standards of the 
 flower are, as in Junos, 
 not erect, but spread out 
 horizontally, or rather 
 turned downwards ; but, 
 unlike other Junos, so 
 far from being minute, 
 or even small, they are 
 relatively as large as in 
 the Xiphium group. In 
 the fall the claw has no 
 lateral wings, but is 
 narrow, suddenly ex- 
 panding into a broad 
 blade ; and the crests 
 of the styles are of 
 moderate size only. In 
 
 all these points I. Fosteriana approaches the Xiphium group. 
 And in correspondence with these features the bulb (fig. 27) is 
 peculiar ; it is thin and slender, covered with several membra- 
 nous olive-green wrappings, and the fleshy roots so characteristic 
 of the Juno group are very feebly developed ; they do exist, but 
 
 FIG. 26. IKIS FOSTERIANA. 
 
45 
 
 they are thin, and are hardly more than somewhat thickened, 
 more persistent, ordinary annual roots. Thus the plant by its 
 several characteristics is intermediate between the Xiphium and 
 the Juno groups, being, on the whole, nearer to the latter ; if 
 we suppose that it has descended from some ancestor more or 
 less allied to /. Sisyrinchium, we may 
 imagine that it has wavered between 
 two lines of development, doubting 
 whether to become a Xiphium or a 
 Juno. 
 
 The flower is not very large, about 
 the size of a small I. xiphium, and 
 its chief merit lies in the colouring, 
 though the form is not without grace. 
 While the falls and the styles are 
 yellow, a rich yellow in some speci- 
 mens, a more or less greenish yellow 
 in others, the turned-down standards 
 are of a full rich purple, and the con- 
 trast between these two hues produces 
 an effect which, though the plant bears 
 my name, I think I may say is very 
 pleasing. I have not as yet perceived 
 any fragrance. 
 
 It does not take kindly to our Eng- 
 lish climate. The leaves often begin 
 to spear in late autumn, and suffer 
 from the buffetings of winter ; it 
 flowers in March, when its slender 
 stem is laid low by fierce winds, and, II 
 judged from the climate of its native" 
 
 home, it needs, what it cannot get ,., 
 
 FIG. 27. BULB OF IRIS FOSTERIANA. 
 with us, the rest of a thorough 
 
 drought in summer. Not possessing the thick fleshy roots of 
 the other Ju?tos, it is less amenable to annual "lifting" than 
 are they; indeed my experience leads me to think that it 
 resents being moved at all. In fact, I find it a very difficult 
 plant to grow, and I doubt if it will ever become common in our 
 gardens. At home, in Gulran, it grows at an altitude of about 
 4,000 feet, in dry places, in what Dr. Aitchison calls " sandy 
 
46 
 
 clay, "and my own experience indicates a moderately stiff, rather 
 than a very light soil, as proper for it. In any case, it should 
 have a spot as dry as possible in winter, and as hot as may be in 
 summer ; it must be sheltered from winds, and should be kept 
 free from the encroaching roots of other plants, and especially 
 of shrubs and trees, for if it is to live at all in this country, its 
 struggle for existence must be made as light as possible. 
 
 All the Junos of which I have so far spoken come, like the 
 wise men, from the East. The centre of their geographical dis- 
 tribution lies in Persia ; they disappear further to the east in 
 Afghanistan and in the Punjaub, and to the west they are lost in 
 Asia Minor and in Palestine ; the most western representatives 
 are, so far as we know at present, those varieties of I. persica 
 which are found in Armenia and Cappadocia. There is, how- 
 ever, a distinctly western member of the group, one only, which 
 flourishes in the extreme west, in Spain, Morocco, and Algiers, 
 and, stretching eastward, is lost in Greece ; so far as I know, it has 
 never crossed the Bosphorus eastward, just as no eastern member 
 of the group has crossed the same strait westward. This solitary 
 western Juno is the Iris, known long ago to the old gardeners, 
 spoken of by Parkinson as " Clusius his first great bulbous flower- 
 de-luce," and called nowadays I. alata (fig. 28), I. scorpioides, 
 and by various other names. It is fairly abundant in Southern 
 Spain and Portugal, in Algiers and Morocco. It is found in Sicily 
 and Sardinia, and though apparently absent from Southern Italy, 
 reappears, as I have said, in Greece. 
 
 It has all the characters of a Juno Iris ; indeed, the group 
 was founded upon it. The bulb has the distinctive fleshy roots, 
 and the claw of the fall has the characteristic wings ; the 
 standards are minute and turned down ; the crests of the style 
 are large and prominent, and the leaves are numerouife, broad, 
 and lax. The prevailing colour is lilac or blue, the conspicuous, 
 generally fringed or laciniate, ridge on the fall being yellow ; but 
 the flower varies very much in tint, and in the prominence of 
 veins of a deeper blue, and of spots or blotches of yellow ; and a 
 white albino variety is in cultivation. The flower also varies 
 greatly in size, some flowers, notably the variety " magna " and 
 one recently distributed by Mr. Max Leichtlin, being of great size 
 and very beautiful ; it is in every way a desirable plant, even the 
 smaller and less highly coloured ones being welcome. It has a 
 
47 
 
 FIG. 28. IBIS ALATA (Caparn.) 
 
48 
 
 very distinct, indeed powerful odour, which, however, to some 
 persons is not agreeable. 
 
 It is distinctly a winter-flowering plant ; in this country it 
 sends up its bloom (which is often single, though sometimes two 
 or even more flowers appear) soon after the leaves begin to push 
 in October, or later, according to climate and situation. It 
 makes most of its growth in winter, the foliage beginning to die 
 off in April or May. Hence, in all the northern parts of England 
 at least, it cannot be considered hardy in the absence of pro- 
 tection ; the frosts, snow, and winds of winter play havoc with 
 its broad lax leaves, and, left to itself, the plant soon disappears. 
 In the milder southern counties it will thrive, and is there justly 
 appreciated. In my own bleak Eastern Counties garden I find 
 it difficult to grow, even with the protection of a frame, in the 
 absence of artificial heat. But it is a plant which repays some 
 little trouble. 
 
 Well-grown and well-ripened bulbs of I. alata, potted in 
 summer, plunged in cocoa-nut fibre or ashes until they begin to 
 grow, and brought into a cool greenhouse as soon as the nights 
 threaten to become frosty, will flower freely ; but, as a rule, for 
 the first year only. Even when great care is taken to keep 
 them suitably supplied with moisture during and after flowering, 
 and to ripen the bulbs properly by the gradual withholding of 
 water and by full exposure to the sun as soon as their growth 
 is completed, the bloom is uncertain at the second year, and the 
 bulbs soon dwindle. This, at least, is my experience, and I 
 believe that of others ; and the result is the same whether 
 the bulbs be left in the pots or be potted afresh year by year. 
 And the same may, I believe, be said of all the Juno group ; 
 bulbs taken from the open will flower in pots the. first year, 
 but rarely beyond that. To keep up a supply in pelts for the 
 greenhouse recourse must be had yearly to a stock grown in 
 the open. 
 
 When it is desired to increase the stock of a Juno Iris, the 
 small bulbs which appear round the base of the larger bulbs 
 may be separated and planted separately, or a mass of bulbs 
 which has arisen from an original bulb may be divided into 
 its several constituents. In doing this care should be taken 
 to preserve as much as possible the fleshy roots ; but their 
 preservation, though desirable, is not necessary ; bulbs which 
 
49 
 
 have lost all their roots will grow, though they are, of course, 
 handicapped by the loss. 
 
 Most of the Juno Irises go 
 to seed very freely, especially 
 I. caucasica, I. Rosenbachiana, 
 and, in an adequately genial 
 climate, I. alata ; and the seed 
 germinates readily. The seed 
 of the two former, since it does 
 not sprout until spring, may 
 be sown in the open, in pre- 
 pared beds, and left there for 
 three or four years, at which 
 time the seedlings should be 
 moved when the 
 foliage has gone 
 down, since many 
 of them will by 
 that time have 
 become flowering 
 bulbs. It is, per- 
 haps, hardly neces- 
 sary to say that 
 when seed is thus sown in the 
 open, it should be covered more 
 deeply, with fine sifted soil, 
 than when it is sown in pots ; 
 otherwise much of the seed 
 will come to the surface and be 
 lost. The seed of I. alata will, 
 of course, need to be sown 
 under protection, and that of 
 I. persica will benefit by a 
 similar treatment. 
 
 Iris nepalensis. 
 
 With I. alata my tale of 
 bulbous Irises comes to an end ; Fl - 29. IRIS NEPALENSIS (reduced one-fourth) 
 and yet I should like to say a word or two about a very peculiar 
 
 D2 
 
50 
 
 Iris, not a bulbous Iris properly so called, but one, nevertheless, 
 which presents certain striking affinities to the Juno group. 
 As I said, the Juno group, so far as we know at present, stops 
 eastward at Afghanistan and the Punjaub. Still further east, 
 in the kingdom of Nepaul, is found an 
 Iris, unlike any other Iris, the I. nepa- 
 lensis of Don (figs. 29 and 30), the 
 I. decora of Wallich (the 
 I. nepalensis of Wallich is 
 quite a different plant, being 
 merely a variety of /. ger- 
 manicd). In its resting 
 condition this Iris consists 
 of a small bud, wholly 
 hidden and covered up by 
 the fibrous remains of the old leaves (fig. 31), 
 and from the base of the bud hang down 
 a number of fleshy roots, very much like 
 those of a Juno Iris, only more numerous, 
 narrower, more thonglike, and more uni- 
 form in size. The whole root, which in a 
 way recalls that of a Hemerocallis, may be 
 compared to that of a Juno Iris, the bases of 
 the old leaves of which, instead of forming 
 the carefully inwrapped membranous coats 
 and fleshy tunics which make up the bulb 
 surrounding the central bud, have been 
 broken up into an apparently irregular nest 
 of fibres and bristles. This analogy in 
 regard to the resting root between the Iris 
 of which I am speaking and a Juno Iris is, 
 moreover, carried into the flower. In J. 
 nepalensis, as in a Juno Iris, the standards, 
 though they are not diminished in size, 
 FIG. 30. IRIS NEPALENSIS. tend to spread horizontally ; the fall bears 
 a very conspicuous median ridge, toothed or 
 even cut up into a beard, and the crests of the style are largely 
 developed. Moreover, as in most Juno Irises, the flower appears 
 while the leaves are as yet young and short. So that, not only 
 in its root but in its flower, I. nepalensis comes nearer to a Juno 
 
51 
 
 Iris than to any other Iris, and I venture to think that we may 
 probably regard it as an Iris which has descended from the same 
 ancestors as the Juno Irises, though it has taken another line of 
 development ; its greatest divergence is shown in the stem some- 
 times being branched. 
 
 The flowers of I. nepalensis, borne on a stem a foot or so in 
 height, are very pleasing, owing partly to their graceful form 
 and partly to their delicate colour, which is a pale lavender, due 
 to delicate violet veins on a creamy white ground. Unfortu- 
 nately, the bloom is very fugitive ; opening in the morning, it closes 
 and is over before evening ; moreover, the plant, in my hands at 
 least, is not very floriferous, and, in addition, needs very special 
 treatment. It has to be supplied with abun- 
 dant moisture all the summer, and yet to be 
 kept very dry all the winter. The latter end 
 may be gained by lifting the roots in late 
 autumn and storing them until spring in dry 
 sand, treating them in fact like bulbs ; but I 
 find that plants so treated are not so vigorous 
 as those which are left in the 
 ground and dried by being 
 covered with lights. To the 
 ordinary gardener the result 
 will not seem worth the 
 trouble, and I do not think 
 that I. nepalensis will ever 
 become a favourite. More is 
 to be hoped from a variety of 
 the species found by Lieut. - 
 Col. Stone on the Chin Hills 
 in Upper Burmah, and sent 
 by him to Mr. P. Barr ; this I 
 have ventured to call I. nepalensis var. Letha. It hardly differs 
 at all from the type, except that the flowers are quite sessile ; 
 but it is much more floriferous, and, so far as my short experience 
 goes, much less impatient of the vagaries of our English climate. 
 A plant, of which I simply helped the ripening by covering it 
 with a light in late autumn, stood exposed and unprotected, save 
 for a loose mulching, all last winter (1891-92), and flowered 
 freely at the end of June. The delicate lavender flowers, nestling 
 
 FIG. 31. BESTING BUD OF IBIS NEPALENSIS. 
 
52 
 
 at the base of a tuft of short green leaves, present a very pleasing 
 sight, and I can well imagine that a mass of the plant would be 
 extremely charming. The flower, moreover, has the merit of 
 being deliciously fragrant. When an adequate stock of the 
 plants has been obtained, it will, I venture to think, become a 
 great favourite. 
 
 And now my tale is quite ended. Some of the flowers of 
 which I have spoken are of extreme beauty ; to my mind at least, 
 /. reticulata and I. Bosenbachiana, in their best forms, present- 
 ing two different types of beauty, are surpassed by few other 
 flowers, and many of the others, especially I. alata, are very 
 charming. But it is obvious from what I have said that in 
 ordinary gardens these bulbous Irises, with the exception perhaps 
 of /. xiphium, I. xiphioides, and I. orchioides, cannot be grown 
 successfully without special care. They cannot be left alone to 
 take care of themselves ; in the struggle for existence they make, 
 as a rule, but a poor fight. They cannot combat with vigorous 
 perennials, or withstand the inroads of the pushing roots of 
 trees and shrubs. If they are to be grown well, they must have 
 a place for themselves, and in most cases their special wants 
 must be carefully supplied. 
 
53 
 
 LIST OF SPECIES. 
 
 In the following pages I give a more formal account of the 
 several species and varieties of bulbous Irises spoken of in the 
 lecture ; but the account in each case is intended for gardening 
 rather than for botanical purposes. I give various synonyms, 
 with the respective authorities, since these are often needed by 
 gardeners ; but I have not attempted to make the list absolutely 
 complete. I give references to figures, since these are often 
 sought for by gardeners. The descriptions under the heading 
 "Characters" are not intended to be botanical descriptions ; I 
 have contented myself with such salient features as will be of 
 help to the gardener. Under the heading " Time " is given the 
 approximate date at which the plant flowers in England. And 
 I have thought that some statement as to the geographical dis- 
 tribution of each kind would be useful and profitable to the 
 gardener. 
 
 At the suggestion of the Secretary, I have added a " key " 
 to the several species and varieties. I do this with great diffi- 
 dence, since it is almost impossible to arrange species in a key 
 under short headings, without bringing into close proximity 
 species which are naturally widely separate ; the arrangement 
 selected is almost sure to be artificial, and so far misleading. 
 Moreover, I feel great hesitation in giving a definite position to 
 any plant which I have not had under cultivation ; and this, to 
 make the key complete, I have been obliged to do. However, 
 such a key as follows may be useful, and I give it for what it is 
 worth. 
 
 KEY. 
 
 I. BULBS WITHOUT SUMMER STORE ROOTS 
 
 A. FLOWERS SESSILE. 
 
 1. Leaves linear : PAGE 
 
 Outer tunics of bulb membranous . (7) I. W'mkleri . . .64 
 
 Outer tunics of bulb netted . . (6) I. Kolpakomskyana . 63 
 
 2. Leaves cylindrical, eight-ribbed . . (3) 7. Bakeriana . . 60 
 
 3. Leaves four-sided : 
 
 o. Standards minute . . . (5) I. Danfordice . . 61 
 /3. Standards not minute : 
 
 aa. Claw of fall much narrower 
 
 than blade . . . (4) I. Vartani ... 61 
 
bb. Claw of fall nearly as broad 
 
 as blade .... 
 Leaves of some height when 
 
 the flower expands : 
 Flowers deep violet 
 Flowers red-purple, with 
 conspicuous veins on 
 the sides of the claw 
 of the fall . 
 Flowers bluish, spotted . 
 Flowers dwarf, red-purple 
 Flowers dwarf, bright blue 
 Leaves very short when the 
 
 flower expands : 
 Segments long and narrow 
 Segments broad : 
 Colour bluish, often 
 
 spotted 
 Colour red-purple 
 
 (2) 1. retimtlata 
 
 I. ret. type 
 
 1. ret. v. Krelagei 
 I. ret. v. Histrio . 
 I. ret. v. humilis . 
 I. ret. v. cyanea . 
 
 I'AGE 
 
 56 
 
 56 
 
 57 
 57 
 59 
 
 I, ret. v. sophenensis . 58 
 
 I. ret. v. histrioides 
 1. ret. v. purpurea 
 
 59 
 
 60 
 
 B. FLOWERS BORNE ON A STEM. 
 
 (1)7. Sisyrinchium . .55 
 /. Sysyr. type . .55 
 I. Sisyr. v. monopJiylla 56 
 
 I. Sysyr. v. maricoides . 56 
 
 (57 
 
 1. Leaves not more than two. Bulb 
 
 globose, with shaggy coats . 
 Anthers adherent below, leaves two 
 
 leaf one 
 Anthers not or rarely adherent. 
 
 Flowers much spotted . 
 
 2. Leaves more than two : 
 
 a. Perianth tube absent : 
 
 o. Segments lax and broad . (10) I. xipliioides 
 j8. Segments stiff and narrow : 
 aa. Fall fiddle-shaped : 
 
 Flowers bluish or white, 
 with segments spread 
 horizontally 
 Flowers yellow, seg- 
 ments forming a fun- 
 nel . 
 bb. Fall not fiddle-shaped : 
 
 Claw of fall nearly as 
 
 broad as blade 
 Claw very narrow 
 
 b. Perianth tube variable. Fall 
 
 fiddle-shaped 
 
 (8) 2. xipUiwn 
 
 I. xipli. v. lusitanica . 65 
 
 I. xipJi. v. Battandieri 
 (9) I. serotina . A 
 
 . (11) I.filifolia . 
 c. Perianth tube distinct and con- 
 spicuous : 
 a. Perianth tube not longer 
 
 than ovary : 
 
 aa. (?) Fall fiddle-shaped . (13) I. Fontanesii 
 bb. Fall with narrow claw 
 
 and ovate blade . (12) I. tingitana 
 ft. Perianth tube much longer 
 
 than ovary : 
 aa. Flower purple, with a 
 
 slight beard on the fall (15) I. Boissieri . 
 bb. Flower yellow . . (14) I. juncea 
 
 65 
 
 71 
 70 
 
 72 
 71 
 
55 
 
 (17) 7. persica type 
 
 I. persica v. purpurea 
 
 I. persica v. Isaacsoni. 76 
 
 IL BULBS PROVIDED WITH SUMMER STORE ROOTS. 
 A. FALL WITH LATERAL WINGS TO THE CLAW. 
 
 1 . Flowers sessile, that is to say, stem not obvious : 
 
 a. Flowers appearing soon after the leaves begin to shoot : 
 
 Flower with conspicuous patch on the blade of the fall : 
 
 Flower, except for patch, 
 nearly white 
 
 Flower purple 
 Flower without conspicuous 
 
 patch ..... 
 
 b. Flowers not appearing until the 
 
 leaves have acquired some 
 
 length : 
 
 o. Flowers greenish yellow, 
 
 small .... 
 
 /8. Flowers blue or purple, large 
 
 2. Flowers borne on a stem more or less 
 
 obvious : 
 
 a. Leaves with conspicuous horny 
 
 margin : 
 
 o. Flowers purple . 
 /3. Flowers yellow : 
 
 Stem hardly visible . 
 Stem obvious : 
 
 Wings to claw conspicu- 
 ous . . . /. caucasica v. major (turkestanica) 
 Wings to claw not con- 
 spicuous 
 
 b. Leaves without conspicuous 
 
 horny margin : 
 a. Leaves broad : 
 
 aa. Flowers bluish . 
 
 bb. Flowers smoky yellow 
 
 cc. Flowers wholly yellow 
 
 dd. Flowers yellow and blue 
 
 te. Flowers yellow, spotted 
 
 with blue 
 . Leaves very narrow 
 
 (20) 1. Palestina. 
 (28) 7. alata 
 
 (23) 7. Stocteii . 
 (18) I. caucasica type. 
 
 (21 ) I. sindjarensis 
 
 (22) l.fumosa . 
 (19) 1. orchioides 
 
 J. orchioides v. ccerulea . 
 
 I. orchundes v. ocidata 
 (?) 7. orchioides v. linifolia 
 
 B. FALL WITHOUT LATERAL WINGS TO THE CLAW. 
 
 1. Leaves broad. Flowers sessile . 
 
 2. Leaves narrow. Stem obvious : 
 
 a. Leaves much arched . 
 
 b. Leaves erect : 
 
 a. Standards large . 
 /3. Standards small . 
 III. RHIZOME A TUBER . 
 IV. RHIZOME A BUD covered with 
 bristles, and bearing summer 
 store roots : 
 
 Flowers borne on an obvious stem 
 Flowers sessile 
 
 (26) 7. ftosenbachiana 
 (25) 7. drepanopliylla . 
 
 (27) I. Fosteriana 
 (24) I. Aitchisoni 
 (16) I. tuber osa . 
 
 74 
 75 
 
 79 
 
 82 
 
 80 
 
 77 
 
 77 
 
 I. caucasica v. Kharput 78 
 
 80 
 80 
 78 
 78 
 
 79 
 79 
 
 81 
 
 81 
 
 82 
 81 
 73 
 
 (29) 7. nepalemis . . 83 
 I. nepalensi* v. LetJia . 84 
 
 (From 
 
 1. IRIS SISYRINCHIUM. Linmjeus, Sp. Plant. 59. 
 the Greek word sisyra, shaggy coat.) 
 
 Synonyms./, fugax (Tenore, Fl. Nap. i. p. 15, t. 4). 7. 
 segyptia (Delile, Ann. Sc. Nat. 1843, ii. p. 237). J. maricoide* (Regel, 
 
56 
 
 St 
 
 Descr. ix. p. 36). J. monophylla (Boissier and Heldreich, Heb. 
 Norm. 51). I. Samaritana (Heldreich). Morsea Sisyrinchium (Ker, 
 Bot. Mag. t. 1407). M. Tenoreana (Sweet, Brit. Flow. Gard. t. 110). 
 M. sicula (Todaro, Hort. Bot. Panorm. t. 34). J. juncifolia, and 
 others. The " Spanish nut " and the " Barbary nut " of Parkinson. 
 
 Figures. Sibthorp and Smith, Flor. Grxca, t. 42 ; Redoute', Lil. 
 t. 29 and 458 ; Bot. Mag. t. 6096, and as above ; Tenore and Sweet, 
 as above. 
 
 Characters. Bulb globose, with shaggy coats. Leaves two, linear, 
 pointed, arched. Stem from a few inches to a foot or more in length, 
 bearing a terminal, and one, two, or more lateral sessile buds, each 
 containing one, two, three, or more flowers. Spathe-valves long, 
 lanceolate, more or less scariose. Upper limit of ovary not sharply 
 defined by any constriction or other outward mark from the perianth 
 tube above. Flower small, lasting when open for a few hours only. 
 Colour varying from light blue to a reddish purple, with variable spots 
 and veins on the blade of the fall, which 
 bears a broad patch (signal) of white, with 
 a median yellow streak, often spotted. 
 Filaments of anthers adherent to the 
 styles and to each other to a variable ex- 
 tent in their lower part. Capsule covered 
 by scarious spathe-valves, rounded-trigo- 
 nal, long, and narrow, with numerous small 
 oval seeds. Flowers often very fragrant. 
 Habitat. Spain, Portugal, Morocco, 
 and thence eastward along both shores of 
 the Mediterranean, Algiers, Italy, Corsica, 
 Sardinia, Sicily, Malta, Tunis, Egypt, 
 Greece and Grecian Islands, Palestine, 
 Fm. 32. IBIS SISYBINCHIUM.* Syria to Asia Minor, Persia, Beloochi- 
 
 stan, Afghanistan, Bokhara, Punjaub. 
 Time. Flowers in this country in June. 
 
 Varieties. An albino white variety is sometimes found. The 
 variety monophylla (I. monophylla of Boissier) from Greece nd Egypt 
 has one leaf only, a stem 2 to 4 inches long, often with a terminal 
 bud only, and small dull-coloured flowers. The variety maricoides 
 (I. maricoides of Regel) from Bokhara has much-spotted flowers, with 
 the anthers for the most part free. 
 
 2. I. RETICULATA. Marschal v. Bieberstein, Fl. Taur. 
 Cauc. i. p. 34. (From the netted, reticulate coats of the bulb.) 
 
 Fig. M. v. Bieb. Cent. t. 11 ; Regel, Gartenflora, t. 677 ; Bot. 
 Mag. 5577 (not quite typical) ; Garden, vol. xx. (1881), p. 112. 
 
 Char act. Bulb ovoid, with netted coats. Leaf four-sided, 
 
 * In this and succeeding figures are given outlines of /. fall, s. standard, 
 st. style, each spread out flat, and of natural size. 
 
57 
 
 FIG. 33. IBIS RETICULATA. 
 
 armed with a horny point. Leaves and flowers with separate vaginal 
 sheaths. Flower single, sessile, with long tube. Leaves of some length 
 when the flower opens. Spathe- 
 valves green. Fall long, narrow, 
 strap-shaped, with hardly any dis- 
 tinction between claw and blade, of 
 a rich deep violet colour, except that 
 the hind part of the blade bears a 
 low, rounded, bright orange median 
 ridge in the middle of a white patch 
 streaked with violet ; the ridge is 
 replaced along the claw by a yellow 
 streak, marked with dots. The 
 median region of the claw on each 
 side of the streak is marked by in- 
 distinct veins which are lost at the 
 margin in the uniform violet ground- 
 colour. Ripe capsule long, narrow, 
 pointed at each end, borne up above 
 the ground by a peduncle. Very 
 fragrant, odour of violets. 
 
 Time. March, but variable. 
 
 Hab. Occurs sparsely in the Caucasus and surrounding regions. 
 
 I. RETICULATA var. KRELAGEI. Regel, Gartenflora, 
 1873, p. 354. (After Krelage, the Dutch horticulturist.) 
 
 Fig. Loddiges Bot. Cab. 1. 1829; Sweet, Brit. Flow. Gard. 2ndser. 
 ii. t. 189 ; Regel, Gartenflora, 1873, t. 779, fig. 2. 
 
 Charact. Flower shorter and broader in all its parts, and tube 
 shorter than in the type. Blade of fall generally marked off from claw 
 by a slight constriction. Veins on claw of fall very distinct, separate 
 up to the margin. Prevailing colour a red-purple, varying from dingy 
 plum-colour to deep (almost black) purple. Ripe capsule short, broad, 
 and blunt, half buried in soil. Different plants vary much in size, in 
 the length of the tube, in the relative breadth of the fall, the exact 
 shape of the blade of the fall, which in some is narrower, longer, more 
 pointed than in others, and in the shape and character of edge, whether 
 serrated or not, of the crests of the styles. Sometimes, but rarely, 
 fragrant. 
 
 Time. February or March ; variable, but generally somewhat 
 earlier than the type. 
 
 Hab. Common in the Caucasus, and spreading into Persia and 
 Asia Minor. 
 
 I. RETICULATA var. HISTRIO. I. Histrio, Reichenbach 
 fil. Bot. Zeit. 1872, p. 388. (From histrio, a stage-player, as if 
 painted for the stage.) 
 
58 
 
 Pig. Bot. Mag. 6033 ; Garden, vol. xxxiii. (1888), p. 558. 
 
 Charact. Leaves of considerable length, a foot, or even more, 
 
 when the flower opens. Tube rather short, and segments somewhat 
 
 short and broad, as in var. Krdagei. Spathe-valves narrow, pointed, 
 
 f ~ nearly colourless and transparent. 
 
 Style more triangular than in 
 /u /\ type ; crests of style larger, more 
 I M / \ triangular and pointed. Differs 
 ] \ / \from the type entirely in colour ; 
 
 / V jthe margin of the blade of the 
 
 I .^jK. / ^ a ^ * s a r *k Bright colour, which 
 \f^ ^'N' is almost a blue, but has a slight 
 1 / admixture of red ; within this 
 
 \ I is a region of a creamy white 
 
 \ I ground with many bright blue 
 
 \ I blotches and broken veins of a 
 
 \ I similar colour ; in the median line 
 
 \ I is a low narrow yellow ridge, 
 I I which is continued on along the 
 
 claw as a yellow streak, dotted 
 
 FIG. 31-lRis BETICULATA var. HISTBIO. with low black tubercles. The 
 
 lateral regions of the claw bear 
 
 veins which at the base of the claw break up into dots. Standards 
 and styles of a lighter blue, the crests of the styles being marked 
 with deeper veins. 
 
 Time. December to March, according to situation and season ; 
 the earliest reticulata. 
 
 Hab. Palestine, Mount Lebanon, Mount Gerizim. 
 
 I. BETICULATA var. SOPHENENSIS. Foster, Gard. 
 Chron. 1885, i. 470. (From Sophene, 
 the ancient name of the district round 
 Kharput.) 
 
 Charact. The flower, with the spathe- 
 valves, appears while the leaves have 
 hardly pierced the ground. Tube v variable 
 in length, but always longer than the 
 spathe-valves. Whole flower more spread- 
 ing, less funnel-shaped than in type. The 
 long, narrow claw of the fall expands into 
 a blade which is only about half as long as, 
 but nearly as broad again as itself. The 
 orange ridge on the blade is continued as 
 a yellow uneven ridge along the whole 
 length of the claw. The colour, which has 
 a peculiar metallic sheen, varies from a 
 dark red-purple to a blue-purple, or to a lilac or lavender. The flower 
 
 FIG. 35. IRIS RETICULATA 
 var. SOPHENENSIS. 
 
59 
 
 is not marked with veins except on the claw, and occasionally on the 
 blade of the fall. 
 
 Time. February ; succeeds Histrio, and precedes var. Krelagei. 
 
 Hab. Asia Minor, nearKharput, and near Mardin. 
 
 I. RETICULATA var. HUMILIS (lowly). 
 
 Charact. Flower appearing while the leaves are a few inches 
 in height. Tube rarely longer than the green spathe-valves, the flower 
 being about two inches high, and more compact than in the other 
 forms. The narrow claw passes, with a 
 slight constriction, into the much broader, 
 ovate-lanceolate blade. The blade of the 
 fall bears a bright orange low ridge, con- 
 tinued along the claw as an orange or yellow 
 streak. Around the ridge on the blade and 
 the streak on the claw is a zone of opaque 
 ereamy- white broken up by dots, and by 
 broken veins of dark purple. The rest of 
 the blade is of a full, rich, red-purple colour, 
 on which the veins are hardly visible, con- 
 trasting strongly with the creamy-white zone ; 
 at the margin of the claw the veins fuse into 
 a like rich red-purple ground colour. The 
 flower in some respects resembles that of 
 I. Bakeriana. 
 
 The variety cultivated under the name I. reticulata var. cyanea 
 agrees with the above, save that the colour is a rich cyanic blue and 
 the dots give place to veins ; of this again a pale blue form exists. 
 
 Time. March, between var. sophenensis and var. Krelagei. 
 
 Hab. Asia Minor, near Van. 
 
 I. RETICULATA var. HISTRIOIDES. (Like Histrio.) 
 
 Fig. Garden, vol. xlii. (1892), p. 364. 
 
 Charact. The flower appears before the leaves, which are even- 
 tually stouter and longer than in any other form of reticulata. Flower 
 larger, four or even five inches across, with a tube variable in length, but 
 always longer than the spathe-valves, which are short,' broad, blunt, and 
 colourless, or nearly so, being marked by a few green veins. Claw 
 of fall not strap-shaped, but triangular, gradually expanding, a 
 slight constriction, however, marking the transition, into a broad ovate 
 blade as long as or even longer than itself. The falls spread hori- 
 zontally, so that the whole flower is less funnel-shaped than is the 
 type. Style triangular, short, with large, more or less quadrangular 
 crests. The narrow orange median ridge on the blade is continued 
 along the claw as a ridge, with an uneven tuberculate surface. The 
 prevailing colour is a bright blue, assuming in some specimens a light 
 
 FIG. 36. IRIS BETICULATA 
 var. HUMILIS. 
 
60 
 
 FIG. 37. IBIS BETICULATA var. HISTBIOIDES 
 
 violet hue. In some specimens the whole blade and claw outside the 
 ridge is of a uniform blue, broken only by veins of a deeper tint ; in 
 g . others the margin only of the 
 
 fall is so coloured, and the 
 region between the margin 
 and the median ridge is, as 
 in Histrio, of a creamy white 
 broken by irregular dots and 
 blotches of blue and by im- 
 perfect veins ; and interme- 
 diate forms occur. The rela- 
 tively large standards and 
 styles are of a nearly uni- 
 form blue. The ripe capsule, 
 short and blunted, is half, or 
 in some specimens wholly, 
 buried in the soil. In a warm 
 atmosphere the flower is dis- 
 tinctly fragrant. 
 
 Time. March ; but vari- 
 able, sometimes as early as sophenensis, sometimes as late as or later 
 than Krelagei. 
 
 Hab. Armenia, district of Amasia. 
 
 I. BETICULATA var. PUBPUBEA, Max Leichtlin. 
 
 Charact. Leaves two or three inches high, or less, at flowering. 
 Tube short, hardly longer than the spathe-valves. The median ridge of 
 the fall, unlike the case of var. Krelagei^ continued all along the claw. 
 Colour a fine deep red-purple, especially dark on the blade of the fall. 
 The sides of the claw of a uniform colour, and not broken up into veins 
 as in var. Krelagei. The flower is intermediate between Krelagei and 
 sophenensis. Not distinctly fragrant. 
 
 Time. March, at about the same time as var. Krelagei. 
 
 Hab. Asia Minor, neighbourhood of Egin. 
 
 I. BETICULATA. Varieties arising in cultivation. , 
 
 The late Rev. Mr. Nelson raised a form which resembles" the type 
 in every way, except that the flower is larger. This is sometimes 
 spoken of as I. reticulata var. major. He also raised some seedlings of 
 the form of the type, but in which the deep violet is replaced by a 
 slight, somewhat slaty blue ; this is sometimes spoken of as I. reticulata 
 var. cserulea. Mr. Max Leichtlin has since raised seedlings similar 
 in form, of different shades of light blue, one of which is of a very 
 large size. 
 
 3. I. BAKEBIANA. Foster, Bot. Mag. vol. xlv. No. 539 
 (Nov. 1889), t. 7084. (After Mr. Baker.) 
 
 Pig. As above. Garden, vol. xxxvii. (1890), p. 462. 
 
Gl 
 
 FIG. 38. IRIS BAKERIANA. 
 
 Charact. Bulb ovoid, with netted coats. Leaves nearly a foot 
 long at flowering. Leaf armed with horny point, as in J. reticulata, 
 but cylindrical, with eight ribs, not four- 
 sided. Flower two to three inches 
 across. Median ridge of fall flattened 
 down, and reduced, so as to be hardly 
 more than a yellow streak. Marginal 
 regions of blade of fall deep violet ; 
 central portions creamy white or yellow- 
 ish, with numerous deep violet spots or 
 blotches. Claw with conspicuous violet 
 veins on the sides, but in the median 
 part more or less yellow, with violet 
 spots. Standards and crests of style 
 bluish lilac. Ripe capsule pointed, 
 thrown up above the soil by the pe- 
 duncle. Very fragrant. 
 
 Time. February and March ; vari- 
 able, sometimes very early. 
 
 Hab. Armenia, near Mardin. A variety with the parts of the 
 flower more slender, and with an obvious ridge on the fall, occurs near 
 Urmiah, in West Persia. 
 
 4. I. VARTANI. Foster, Gard. Chron. 1885, i. 438. (After 
 Dr. Vartan, of Nazareth.) 
 
 Pig. Bot. Mag. t. 6942. 
 
 Charact. Bulbs with netted coats. 
 Leaf four-sided, armed with horny point, 
 four inches or moie high at flower- 
 ing time. Flower three or four inches 
 across. Claw of fall very narrow, sud- 
 denly expanding into an ovate-lan- 
 ceolate blade, longer and much broader 
 than itself. Median ridge of fall con- 
 spicuous, smooth, and yellow or white 
 on the blade, tuberculate and yellow 
 with black dots on the claw. Crests of 
 style very long, triangular, narrow, 
 pointed. General colour a dull laven- 
 der, produced by numerous lavender or 
 violet veins, more or less confluent, on 
 a creamy-white ground. Not fragrant. 
 
 Time. October to December. 
 
 Hab. Palestine, near Nazareth. 
 
 5. I. DANFORDLS!. Baker, Journ. of Botany, 1876, p. 265. 
 (After Mrs. Danford.) 
 
 \J 
 
 FIG. 39. IBIS VARTANI. 
 
62 
 
 Syn. 1. Bornmiilleri (Haussknecht, Flora, 1889, p. 140). 
 
 Fig. Bot. Mag. t. 7140 ; Garden, xxxvii. (1890), p. 462 ; Garten- 
 flora, 39 (1890), t. 1327. 
 
 Charact. Bulb with netted coat. Leaf four-sided, armed with 
 horny point. The flower, one to two inches across, appears with or even 
 f S St before the leaves. Claw of fall narrow, sud- 
 
 denly expanding into an ovate blade. On the 
 blade is a conspicuous median orange ridge, 
 continued on along the claw. Standard re- 
 duced to a mere bristle, invisible at a little 
 distance. Style short, triangular, with large 
 quadrangular crests. Whole flower funnel- 
 shaped, the blade of the fall spreading hori- 
 zontally, not deflexed. Colour of all the 
 parts a rich yellow, with variable dark brown 
 
 dots on the blade of the fall near the ridge 
 FIG. 40,-lKis DANFOHDI*. and ftlong ^ ^^ 
 
 Time. February ; sometimes even earlier. 
 
 Hab. Asia Minor : Cilician Taurus, near Mount Amascha ; 
 Amasia, Egin. 
 
 NOTE. Mr. Baker, in his original description (loc. ind.) of I. Dan- 
 fordise gathered in the Cilician Taurus, states that the coats of the 
 bulb are membranous. Haussknecht, in his description (loc. ind.} of 
 1. Bornmiilleri, states that the coats are reticulate (fibroso-reticulatis), 
 and by this differentiates it from I. Danfordise. In his " Handbook of 
 Iridese," Mr. Baker, while making I. Bornmiilleri a synonym of Dan- 
 fordise, describes the coats as " membranous." Not only, however, does 
 Haussknecht insist on the reticulate coats of his /. BorwmuUeri, but 
 bulbs gathered by Bornmiiller, which Mr. Max Leichtlin kindly sent 
 .me, so closely resembled bulbs of reticulata, that of the same lot of 
 bulbs labelled by so careful a man as Max Leichtlin as "Born- 
 miilleri," while most proved to be so by the flower and mode of 
 growth, one or two were real reticulatas (I forget the exact variety). 
 Stronger proof of the distinctly reticulated coats of I. Bornmiilleri could 
 hardly be wanted. On the other hand, I possess, through t the great 
 kindness of Mrs. Danford, a dried specimen of the flower of the Iris 
 which she gathered at Amascha ; and I have no doubt as to its identity 
 with the flower of Bornmiiller's Iris gathered in Amasia or in Egin 
 (according to Max Leichtlin the Amasia plants differ somewhat from 
 the Egin plants). It is most extremely unlikely that there should be 
 two Irises with the flowers indistinguishable, but one having mem- 
 branous coats and the other netted coats to the bulb. I am driven to 
 suppose that I. Danfordise from Amascha and I. Bornmiilleri from 
 Amasia are really the same plant, and that Mrs. Danford's bulbs had 
 lost their outer netted coats before they came into Mr. Baker's hands ; 
 the inner coats of all the reticulata group are much less distinctly 
 
68 
 
 netted than are the outer ones. Mr. Baker's description was confessedly 
 founded on imperfect material ; he was thus led to speak of the leaves 
 as terete or subterete, instead of tetragonal. Mr. Baker, who in de- 
 scribing this Iris in the Journal of Botany (loc. cit.) made for it a special 
 group, now (Iridex, p. 44) places it in the Juno group, being led to 
 do so by the minute size of the standards. Mere difference in the 
 size of any part of the flower must, judging the matter from a theo- 
 retical point of view, be of subordinate importance, of little real 
 weight when opposed to other characters. In all respects save the 
 minute standards, in the bulb, in the leaves, in the flower, and in 
 general habit, I. Danfordix agrees with the reticulata group ; in no 
 other respect than the minute standards does it agree with the Juno 
 group. I have, therefore, taken it in connection with the reticulata 
 group. 
 
 6. I. KOLPAKOWSKIANA. Regel, Act. Hort. Petrop. v. 
 263. (After General Kolpakowsky.) 
 
 Pig. Regel, Gartenflora, 1878, t. 939 ; Bot. Mag. t. 6489 ; Garden, 
 xxxiii. (1888), p. 558. 
 
 Charact. Bulb globose, with netted coats, coarser and looser 
 than those of I. reticulata. The flower, single, sessile, two inches or 
 so across, appears while the leaves are a few inches high. The flower 
 and its tuft of leaves are invested in one common vaginal sheaf ; in 
 J. reticulata. the flower and each leaf 
 has its own separate sheath. Leaf linear, 
 with each edge thickened into a ridge, 
 thus making the leaf a narrow flattened 
 channel. Claw of fall narrow, long, ellip- 
 tical, separated by a narrow constric- 
 tion from the broader, longer, lanceolate 
 blade. The blade bears a low orange 
 median ridge, which becomes a me*re 
 streak on the claw. Standards nearly 
 as long and nearly as broad as the fall, 
 with a short claw ; the blade a very 
 elongated ellipse. Crests of style tri- 
 angular, long, narrow and pointed. The 
 styles unite into a column above the 
 tube. Colour of blade of fall a rich 
 red-purple, with deeper veins, the area 
 round the ridge being a creamy white, with broken veins. Claw of 
 fall yellowish, tending to green, with broken purple veins. Standards 
 and styles light purple or lilac. Stigma simple, or at least not dis- 
 tinctly bifid, as it is in all the reticulata group. Flowers fragrant. 
 
 Time. March. 
 
 Hab. Turkestan, Karatau Mountains. 
 
 ^ KoLpiKOW8KIANi 
 
64 
 
 7. I. WINKLERI. Regel, Descr. PL Nov. ix. (1884), p. 37. 
 (After Mr. Winkler, of the St. Petersburg Botanic Garden.) 
 
 Charact. Bulb ovate, with membranous, not netted, coats. 
 Flower single, sessile, with a vaginal sheath common to it and its tuft 
 of leaves. Leaves four to a tuft, linear. Claw of fall triangular, ex- 
 panding into an elliptical blade. Standards erect and narrow. 
 
 Time. Not in cultivation. 
 
 Hab. West Turkestan, near the river Narju. 
 
 8. I. XIPHIUM. Linnseus, Sp. PL 58. (From 
 sword.) 
 
 Syn. Xiphion vulgare (Miller, Diet. ed. 6). The " Iris bulbosa 
 angustifolia " of Parkinson. 
 
 Pig. Bot. Mag. t. 656 ; Redouts', Lil. t. 337 ; Garden, 1881. 
 
 cp 
 
 FIG. 42. IRIS XIPHIUM (ov. ovary, cp. capsule). 
 
 Charact. Bulb ovoid, with smooth membranous brown coats. 
 Leaves, frequently shooting in late autumn, awl-like at first, linear 
 when fully grown. Stem two feet or less in height, partly covered by 
 clasping leaves, bearing two flowers. Spathe-valves long, narrow, 
 not much inflated. The flower, four inches or so across, devoid of 
 distinct perianth tube, is borne above the spathe-valves by a long 
 pedicel, the ovary being in consequence often much exserted. Fall 
 
65 
 
 fiddle- shaped, having an orbicular blade, separated by marked con- 
 striction from the long claw, which either spreads horizontally or 
 forms, with its fellows, a funnel. Standards narrow, erect. Crests 
 of style large, quadrate. Capsule long, narrow ; seed numerous, 
 small, cubical or angular. Colour either a variable hue of purple 
 (with a narrow orange signal on the blade of the fall) or pure yellow, 
 or yellow blotched with purple. 
 
 Time. June. 
 
 Hab. South of France, Spain, Portugal, Algiers, Corsica (?), 
 Sardinia, Riviera. 
 
 I. XIPHIUM var. LUSITANICA. (After Lusitania, 
 Portugal. ) 
 
 Syn. I. lusitanica (Gawler), Ker, Bot. Mag. t. 679. 
 
 Char act. A form in which the flowers are pure yellow, with an 
 orange signal on the fall, and the perianth more or less distinctly 
 funnel-shaped ; the claw of the fall is generally broad, overlapping the 
 style at the sides, and covering in the base of the standard ; the ovary 
 is less exserted than in the type, and the spathe-valves have some 
 tendency to be inflated. The Xiphion sordidum of Salisbury (Trans. 
 Hort. Soc. i. 303) resembles this, save that the flowers are not wholly 
 yellow, but blotched with purple. 
 
 NOTE. The garden form, known as " The Thunderbolt," is a robust 
 form, with a tall stem, two feet or more in height, and large flowers of 
 a peculiar bronzy or smoky hue, due to an admixture of purple and 
 yellow with brown. Besides its robustness and peculiar colour, one of 
 its most marked features is the breadth of the yellow " signal " on the 
 blade of the fall. This (which, however, is seen, though to a less 
 extent, in some other garden forms) and the fact that it rarely, if ever, 
 in my experience, goes to seed, are perhaps the only indications that 
 it is a hybrid ; if it is, the breadth of the signal would indicate 
 J. filifolia as one of the parents ; and further, if so, I. xiphium var. 
 lusitanica is probably the seed-bearer, since it is very different from 
 seedlings which I have raised from I. Jilifolia as seed-bearer crossed 
 with I. xiphium. Though "The Thunderbolt" shows some of the 
 characters of the hisitanica variety, it is wholly different from the plant 
 described by Salisbury (loc. cit.) as Xiphion sordidum. It may be the 
 I. spectabuis of Spach (Hist. Phan. xiii. p. 20), which that author 
 suggests to be a hybrid of J. xiphium and xiphioides ; but, if so, Spach's 
 suggestion is invalid. 
 
 I. XIPHIUM var. BATTANDIERI. (After M. Battan 
 dier, of Algiers.) 
 
 Syn. I. xiphium (Ba,Hamdier, Bull. Soc. Bot. de France, 1884, p. 366). 
 
 Char act. Bulb, foliage, and general habit same as in type, save 
 that leaves are very glaucous. Perianth very turbinate. Claw of fall 
 not elliptical and separated by marked constriction from rounded 
 
66 
 
 blade, but gradually expanding into ovate blade, there being hardly 
 any constriction between the two. Blade somewhat obovate-emar- 
 ginate, with very revolute edges. Standards tall, broad, very much 
 twisted. All parts of the flower a pure dead white, except the bright 
 orange ridge on the fall. 
 
 Hab. Algiers, Marais de la Rassanta. 
 
 FIG. 43. IBIS XIPHIUM var. BATTANDIERI. 
 
 NOTE. M. Battandier loc. cit.) regards this as being truly 
 I. xiphium, the above being the only habitat in Algiers ; he considers 
 the common xiphium-l&e bulbous Iris of Algiers to be I. Fo\tanesii. 
 
 A variety of I. xiphium gathered by Mr. Maw in the Sierra Nevada 
 of Spain, besides being very slender in foliage, differs from the type 
 in ' the blade of the fall being relatively large, and having a distinctly 
 crenate edge ; the signal, of a bright orange, is also more sharply defined 
 than in the type. The plant, however, hardly deserves a distinct name. 
 
 9. I. SEROTINA. Willkomm, Prodr. M. Hisp. i. p. 141. 
 (From serotinus, late.) 
 
 Fig. None. 
 
 Charact. Resembles J. xiphium, but appears to differ not only in 
 its very late flowering, but in being less vigorous, with the upper 
 leaves very thin and awl-like, in having reddish spathe-valves, in the 
 flowers being very much protruded from the spathe-valves, and 
 
67 
 
 especially in the fall having an oblong-lanceolate blade and a narrow 
 linear claw. Were the plants found in good condition ? 
 
 Time. August or September. 
 
 Hab. Spain, on calcareous mountains near Jaen. 
 
 f 
 
 St 
 
 FIG. 44. IBIS XIPHIOEDES. 
 
 10. I. XIPHIOIDES. Ehrhart, Bei^rag. vii. p. 140. (Like 
 xiphium.) 
 
68 
 
 Syn. Xiphion latifolium (Miller, Diet. ed. 6). J. pyrenaica 
 (Bubani, Sched. Grit. p. 3). The " Iris bulbosa major sive Anglica " 
 of Parkinson. 
 
 Pig. Bot. Mag. t. 687 ; Redoute, Lil. t. 212 ; Garden, xxxi. 
 1887), p. 212. 
 
 Char act. Bulbs with brown coats, which," when old, tend to 
 become shaggy by splitting up into fibres at the top. Leaves stouter 
 and broader than in I. xiphium, not appearing until the spring. The 
 stem bears two, sometimes three flowers. Spathe-valves ventricose. 
 Pedicels short, so that the ovary is not, as frequently in I. xiphium, 
 exserted. Blade of fall rounded, with wavy edge, suddenly narrowed 
 to claw, which itself gradually narrows to its attachment. Capsule 
 large, a long pointed oval, with three flattened sides. The seeds, 
 rounded or pyriform, with wrinkled coats, though large, do not fill the 
 cavities of the chambers. Wild specimens are usually of a rich deep 
 blue, with a conspicuous golden signal on the fall. The cultivated 
 varieties range from blue, through various hues of purple, to almost 
 red, and many are variously mottled or blotched. There is also a 
 pure white variety. 
 
 Time. July, succeeding J. xiphium. 
 
 Hab. French and Spanish Pyrenees, stretching in the West of 
 Spain as far as Burbia. 
 
 11. I. FILIFOLIA. Boissier, Voy. Esp.p. 602, t. 170. (From 
 thread-like, filiform leaves.) 
 
 Syn. Xiphion filifolium (Klatt. Linnsea, xxiv. p. 571). "Iris 
 bulbosa Africana serpentrise caule." l< The purple or murrey-coloured 
 bulbous Barbary Flower-de-luce," Parkinson. 
 
 Fig. Bot. Mag. t. 5928, and also t. 5981 ; Boissier, as above. 
 
 Charact. Bulb like that of J. xiphium, except that the outer 
 tunics are more slender, and veined rather than ribbed. Stem from a 
 foot to a foot and a half in height, with foliage like that of I. xiphium, 
 but, as a rule, narrower, and in the typical form quite filiform. 
 Flower like that of a turbinate I. xiphium, save that sometimes at 
 least a distinct perianth tube of nearly an inch in length is present, 
 that the blade of the fall is more ovate, less rounded, the^standard 
 broader, more obovate, with a notched apex, and the whole flower is 
 either of a dark violet or of a rich red-purple, especially full and 
 dark on the blade of the fall, and that the " signal" patch of orange 
 on the blade of the fall is large, and ends broadly and abruptly, not 
 narrowed to a point ; in the median line of the patch is a slight 
 raised ridge. The capsule, like that of J. xiphium, is long, narrow, 
 rounded, trigonal, with deeply grooved sides, often clavate, and the 
 seeds are small, numerous, and wedge-shaped or angular. The 
 sheaths of the first leaves which pierce the ground are spotted or 
 blotched with purple. 
 
 Time. End of June. 
 
69 
 
 Hab. Spain, Sierras de Mijas and Bermeja, Gibraltar, Morocco, 
 Algiers. 
 
 NOTE. Boissier, in founding this species on plants gathered by him 
 in the Sierra de Mijas and Sierra Bermeja, insisted on the extremely 
 filiform leaves, a quarter as broad as those of I. xiphium, on the 
 possession of a perianth tube half an inch or an inch long, and on the 
 colour (a deep blue-purple) of the flower. In his figure not only does 
 the inch-long perianth tube protrude beyond the spathe-valves, but a 
 considerable part of the ovary is also exserted ; and the colour of the 
 flower is a distinctly bluish purple. In the plants gathered at 
 
 FIG. 45 IBIS FILIFOLIA. 
 
 Gibraltar which I have cultivated, the leaves, though narrower than 
 those of I. xiphium, are much more than a quarter as broad, the 
 ovary is wholly covered by the spathe-valves, the perianth tube is very 
 short (\ inch), and the colour is a rich reddish-purple. These plants, 
 which correspond very closely to that figured in Bot. Mag. t. 5928, 
 approach much nearer to I. xiphium than does Boissier's original 
 plant. The plants gathered at Laraisch, about ten miles S. W. of Tangiers, 
 figured in Bot. Mag. t. 5981, though there named J. tingitana, possess 
 a flower which di fliers markedly from /. tingitana, and is practically 
 identical with that of t. 5928 ; yet the foliage, so far from being 
 filiform, is exceedingly luxuriant, exceeding in breadth even that of 
 
70 
 
 I. xiphium ; so much so that the plant almost deserves the contra- 
 dictory name I. filifolia var. latifolia ! 
 
 We may, perhaps, take up the following position. I. xiphium, much 
 as it varies in form, and especially in colour, never possesses a con- 
 spicuous perianth tube ; though it varies in the exact tint of its blue, 
 never has that blue so much mixed with red as to become a 
 typical purple ; has narrow standards, bears a narrow orange " signal " 
 on the fall, and does not vary largely in the breadth of its foliage. 
 I. filifolia, on the other hand, varies much as to the breadth of 
 the foliage and as to the length of the perianth tube, has a broad 
 signal on the fall, has broader standards, and is distinctly purple in 
 colour, the exact hue of the purple being sometimes bluish, some- 
 times reddish. 
 
 12. I. TINGITANA. Boissier 
 (From Tingitana, Morocco.) 
 
 f 
 
 and Reuter, Pugill p. 113. 
 
 St 
 
 FIG. 46. IKIS TINGITANA. 
 
 Syn. Xiphion tingitanum (Baker, Seeman's Journ. 1871, p. 13). 
 Pig. Bot. Mag. t. 6775 ; Garden, vol. xxxvi, p. 294 (1889). 
 Charact. Bulb large, ovoid, pointed ; outer tunics thin, reddish 
 
71 
 
 brown, with conspicuous veins. First (vaginal) leaf often red, some- 
 times spotted. Stem, which bears two flowers, each five or six inches 
 across, about two feet high, completely hidden by the deeply channelled 
 leaves, which are broad, like those of I. xiphioides, but very glaucous 
 and striated on the outside. Spathe-valves long, reaching up to 
 flower, somewhat swollen, keeled, pointed, green at flowering ; a 
 distinct perianth tube, an inch or more long. Fall, with long claw 
 spreading nearly horizontally, separated by slight constriction from 
 the sharply deflexed, elliptical blade, which has a wavy edge, and is 
 notched at the apex. Standard erect, linear-lanceolate, with wavy 
 edge. Style cuneate, with large, rhomboidal, plaited, and veined 
 crests. Anthers long and large, with orange pollen. Colour of fall 
 light or deep blue, or bluish purple, with deeper veins ; the claw 
 bears a very low median yellow ridge, which on the hind part of the 
 blade spreads out into a broad yellow signal. Standards and styles 
 usually deeper in colour than fall, sometimes markedly so. 
 
 Time. March or April. 
 
 Hab. Tangiers. 
 
 13. I. FONTANESII. Grenier et Godron, Fl. de France, iii. 
 245. (After Desfontaines, French botanist.) 
 
 Charact. This name was given by Grenier and Godron to the 
 Algerian Iris described by Desfontaines as J. xiphium. They describe 
 it as being much larger than J. xiphium, and with a more oval blade to 
 the fall. They do not mention nor, indeed, does Desfontaines the 
 existence of an obvious perianth tube above the ovary, which is a 
 conspicuous feature of the specimens labelled Iris Fontanesii in the 
 Kew Herbarium, and which led Mr. Baker (Journ. ofBot. 1871, p. 13) 
 to regard it as identical with I. tingitana. Grenier and Godron, more- 
 over, say thafc I. Fontanesii occurs also in Spain ; and we may, perhaps, 
 infer that these authors really referred to a large form of I. xiphium, 
 as does also, possibly, Battandier (Bull. Soc. Bot. de Fr. 1886) when 
 he says that I. Fontanesii is very common in Algiers ; in fact, Bat- 
 tandier regards what I have just described as xiphium var. Battandieri 
 as the only true I. xiphium growing in Algiers ! Whether the 
 specimens from Algiers, having a distinct tube above the ovary, are 
 merely varieties of I. tingitana, or whether they represent a distinct 
 species, I should not like to say until I have studied living specimens. 
 I may remark that the existence of a perianth tube above the ovary 
 seems a valid specific character. Though seedlings of J. xiphium and 
 J. xiphioides have been raised generation after generation, none of 
 them, as far as I have hitherto seen, possess such a tube. 
 
 Hab. Algiers. 
 
 14. I. JTJNCE A. Desfontaines, Fl. Atlant. i. 39, t. 4. (From 
 juncus, rush-like leaves.) 
 
72 
 
 Syn. I. imberbis (Poiret, Voyag. ii. p. 86). I. mauritanica of 
 Clusius, and also of Crawler (Ker), Bot. Mag. sub tab. 986. 
 
 Fig. Bot. Mag. t. 5890 ; Desfontaines, as above ; Garden. 
 
 Charact. Bulb globose, with thick, woody, rigid, reddish-brown, 
 ribbed outer coats, a number of which surround an old bulb and split 
 up into stiff fibres. Leaves very slender, rush-like ; they shoot in late 
 autumn. Stem bears one or, when in good condition, two flowers. 
 Spathe-valves long, lanceolate. Perianth tube one or two inches 
 long, slender. Blade of fall round, or nearly so, much broader than 
 the short, nearly triangular claw, which is curved sideway into a 
 
 o,v 
 
 FIG. 47. IKIS JUNCEA. 
 
 canal. Standard oblanceolate, shorter than fall. Style triangular, 
 with large quadrate crests raised above the claw of the fall Colour 
 a rich yellow, a very low ridge on the blade of fall being of a deeper, 
 more orange yellow. Capsule very long, narrow, rounded, trigonal, 
 with deeply grooved sides. Seeds numerous, small, wedge-shaped or 
 cubical. 
 
 Time. June or July, later than I. xiphium. 
 
 Hah. Algiers, Tunis, Morocco, Sicily, Italy (Riviera). 
 
 15.1. BOISSIERE. Henriquez, Bolet. da Soc. Brot. iii. p. 183, 
 with fig. (After Boissier, Swiss botanist.) 
 
 Fig. Willkomm, HI. PL Hisp. t. 118; Bot. Mag. t. 7097 ; Hen- 
 riquez, as above. 
 
73 
 
 Charact. Bulb small, with stiff membranous coats. Leaf linear, 
 very channelled, ribbed on outside. Stem about a foot high, almost 
 entirely hidden, except at the top, by three or four clasping leaves, 
 which, filiform and pointed above, are dilated at the base, where they 
 clasp the stem. Flower single, from two to three inches across. 
 Tube nearly two inches long, but almost wholly hidden by the 
 persistently green, somewhat keeled spathe-valves. Fall spreading 
 horizontally or gently arched downwards, so that the perianth does 
 not form a funnel at all. Claw of 
 fall narrow, strap-shaped, sud- 
 denly expanding, after a slight 
 constriction, into a rounded oval 
 blade much broader than itself. 
 Standard erect, obovate, relatively 
 broad. Style narrow, lying close 
 down on the claw of the fall, with 
 crests which seem to vary, being 
 sometimes triangular, sometimes 
 very large and quadrate. The 
 blade of the fall a rich red-purple, 
 except in the centre, where is 
 a narrow oval patch of bright 
 orange, continued backwards 
 some way along the claw. On 
 this * ' signal " patch are a num- 
 ber of thin yellow hairs forming 
 almost a beard, which reaches 
 nearly to the front end of the 
 patch. Standards purple above, 
 reddish below. Styles of a red- 
 purple, the crests being darker 
 and veined. Capsule rounded, 
 trigonal, with deeply grooved 
 sides, an inch or more in length. 
 
 Time. June. 
 
 Hab. Spain, Gerez Mountains. 
 
 FIG. 48. IRIS BOISSIERI. 
 
 16. I. TUBEROSA. Linn. Sp. PL 58 ; Thunberg, Diss. No. 43. 
 (From tuber-like root.) 
 
 f Syn. Hermodactylus tuberosus (Salisbury, Trans. Hort. Soc 
 i. 304). The velvet Flower-de-luce of Parkinson. 
 
 Fig. Bot. Mag. 531 ; Redout^, Lit. t. 48 ; Sibthorp and Smith, 
 Flor. Gruec. t. 41 ; Sweet, Brit. Fl. Gard. 2nd ser. t. 146 ; Reichen- 
 bach, Ic. Fl. Germ. t. 348, fig. 476 ; Fl. des Serres, t. 1083. 
 
 Charact. Rootstock a tuber, which when full-grown is 
 branched. Leaf often very long, four-sided, with a horny point not 
 
74 
 
 OV 
 
 so well developed as in J. reticulata. Flower about two inches across, 
 funnel-shaped, single, on a stem a foot long. Spathe-valves (some- 
 times one only is present, sometimes the inner one is entirely hidden 
 within the outer one) long, pointed, swollen, persistent after flowering. 
 Flower with long peduncle and short (|-inch to -inch) tube. Fall 
 with an ovate blade, half as long as and generally narrower than the 
 oblong-cuneate claw. Standard minute, narrow, elliptical or lanceo- 
 lated, folded up sideways, with the apex prolonged into a cusp or 
 awn. Crests of style long, pointed, triangular. The styles are often 
 united at some distance above the tube. The partitions of the ovary 
 
 do not meet in the middle 
 line, hence the chamber of 
 the ovary is single. Cap- 
 sule obovate, swollen. The 
 blade of the fall is a deep 
 velvety black, the rest of 
 the flower a yellowish 
 green, with usually a bright 
 green median streak on the 
 claw and beginning of the 
 blade of the fall. The pe- 
 rianth segments are united 
 at their bases, above the 
 perianth tube, into a con- 
 spicuous cup. 
 
 Specimens from different 
 localities vary a good deal 
 in the length of leaves, as 
 to the presence of a second 
 spathe-valve, as to the 
 exact form of the fall and 
 of the standard, and to some extent in colouring ; some specimen 
 have a reddish tinge on the claw of the fall. 
 Time. March or April. 
 
 Hab. South of France, Corsica, Riviera, Sicily, Italy^ Algiers 
 and North Africa, Dalmafcia, Istria, Greece, Ionian and other Grecian 
 Islands. 
 
 Sweet (Brit. FL Gard. 2nd ser. t. 146) makes three distinct 
 species : H. lotigifolius, the form with very long leaves figured by him, 
 coming from Naples ; H. repens, with short leaves and a single 
 spathe-valve, figured in Redoute, and Sibthorp and Smith ; H. U- 
 spathaceus, the form with the two spathe-valves, figured in Bot, Mag. 
 
 17. I. PEBSICA. Linn. Syst. p. 79. (From Persia, habitat.) 
 Syn. Xiphium persicum (Miller, Diet. 6th ed.). Iris bulbosa 
 persica (Parkinson, Parad. p. 172). 
 
 FIG. 49. IKIS TUBEEOSA. 
 
75 
 
 St 
 
 Fig. Bot. Mag. t. 1 ; Redoute, Ltt. t. 189 ; Gard&i, vol. xxxiii. 
 (1888), p. 558. 
 
 Charact. Bulb ovoid, sometimes as large as a hen's egg, with 
 fleshy radish-like roots. Leaves four to five to a tuft, linear or 
 linear-lanceolate, variable in breadth in fact, but always narrowing to 
 a point, more or less arched, and bent laterally. Flower sessile, two 
 or three inches across, generally single, appearing soon after the leaves 
 begin to shoot. Tube two to three inches in length, but variable. 
 Claw of fall, which bears a 
 median, generally orange, ridge, 
 is abruptly expanded laterally 
 into triangular wings or flanges, 
 which, curving upwards, em- 
 brace the style ; blade narrower 
 than the winged claw, more 
 or less ovate, always deeply 
 coloured, with a median orange 
 ridge. Standard minute, half 
 an inch long, lanceolate, with 
 toothed or crenate edge, spread- 
 ing horizontally. Crests of 
 style large, conspicuous, quad- 
 rate, with crenate or toothed 
 edge. 
 
 Varies very much in colour, 
 and also to some extent in form. 
 Colour of type pale blueish green, almost white, with a deep violet, 
 almost black patch on the blade of the fall, and numerous dark spots 
 on and around the orange ridge. 
 
 The var. purpurea (Max 
 Leichtlin) is almost wholly 
 dark red-purple, with an 
 orange ridge on the fall, and 
 a deeper tint on the blade of 
 the fall. Other varieties are 
 light purple, almost lavender, 
 yellowish grey, sea-green, &c., 
 but all have a conspicuous 
 patch of deep purple, violet, 
 or even black on the blade of 
 the fall, and an orange or 
 yellow ridge. 
 
 Time, March, but very 
 
 variable, according to particu- V */ 
 
 lar situation and season. FIG. 51. IBIS PBESICA var. PUBPUBEA. 
 
 Hab. Persia, and throughout the more southern parts of 
 Eastern and Middle Asia Minor. 
 
 FIG. 50. IBIS PERSICA. 
 
76 
 
 I. PERSICA var. ISAACSONI. Var. nov. (After Mr. 
 Isaacson.) 
 
 Charact. Bulb having the ordinary characters of a Juno Iris, 
 somewhat elongate, with brown outer tunics. Leaves four to a tuft, 
 each with a colourless sheath, about five inches long and a quarter inch 
 broad at flowering time, but subsequently becoming longer ; very 
 pointed ; extremely falcate, forming almost a semicircle ; extremely 
 channelled and marked, not only with a conspicuous horny margin, 
 but on the outer surface with horny ridges over the parallel veins, 
 giving it a very ribbed and, at the same time, glaucous appearance. 
 Flowers one, two, or possibly more, sessile, among the leaves, each 
 about two inches across ; very fragrant. Spathe- valves an inch or 
 more long, very pointed, not ventricose, not keeled, colourless and 
 transparent, except for a few green veins. Tube rather shorter than 
 the spathe-valves, cylindrical, marked with violet longitudinal stripes. 
 Fall having a rhomboidal claw, with conspicuous wings, which, opaque 
 at first, become membranous later ; and an ovate-lanceolate-emar- 
 ginate blade, which is sharply reflexed on the claw. The blade 
 bears at its hinder part a very low, inconspicuous yellow median ridge, 
 hardly deserving the name, continued on the claw as a yellow streak 
 marked with violet dots. The rest of the fall is a creamy white tinged 
 with green, and broken by thick violet veins, which, running parallel 
 to the median streak on the claw, form a conspicuous violet zone 
 around the ridge on the blade. Standard minute, horizontal or turning 
 downwards, lanceolate, with conspicuous teeth, creamy white with 
 greenish tinge. Style creamy white, with a greenish tinge in the 
 median region ; much narrower than the fall, somewhat expanded 
 below, with large, quadrate, toothed crests, and a somewhat serrate, 
 horizontal stigma. The three styles are united into a column at some 
 distance above the tube, conspicuous in a side view. Stamen with 
 filament longer than anther, and coarse yellow pollen. Ovary and 
 capsule not seen. 
 
 Time. In its native home in February or March. Cultivated 
 specimens flowered in a greenhouse in October or November. 
 
 Hab. South Persia, Tung-i-Turkan, Kotal Kamarij,^ Seenah 
 Safid, on dry open spots in debris of gypsum ; very local, not found 
 above Shiraz. 
 
 NOTE. This Iris is obviously closely allied to I. persica, but differs 
 from the type and varieties in the following points : In the extremely 
 falcate character of the leaves and the ribbed feature due to the horny 
 ridges ; in the spathe-valves being colourless, transparent, neither 
 ventricose nor keeled ; in the lack of development of the median ridge 
 on the fall ; and in the absence of any deep patch of colour towards 
 the apex of the blade, which is so marked in the type, and is seen in 
 every variety of 1. persica so far known to me. 
 
 I am indebted for bulbs of this plant to John Isaacson, Esq. , of the 
 
77 
 
 Telegraph Department, Bushire. Unhappily, all my plants died before 
 I could fully trace out the history ; those planted in the open started 
 in October, and were killed by the frost in winter, and those grown in 
 pots, after flowering, dwindled away. I have hitherto hesitated to 
 describe it, hoping that fresh plants and further knowledge would 
 enable me to decide whether it should be considered a new species or 
 only a variety of I. persica, a point on which I feel in doubt. I have 
 ventured, as the safer plan in the absence of adequate knowledge, to 
 consider it provisionally as a variety of J. persica. 
 
 18. I. CATJCASICA. Hoffinann, Comm. Soc. Phys. Mosc. i. 
 p. 40 ; Marschal v. Bieberstein, Fl. Taur. Cans. i. p. 33. (From 
 Caucasus, habitat.) 
 
 Lit. Baker, Gard. Chron. 1876, i. p. 692 ; Maximo wicz, Act. 
 Hort. Petrap. vi. p. 417 ; Melanges Biologiques (Bull. Acad. Imp. d. 
 Sci. d. St. Petersb.\ x. (1880) p. 688. 
 
 Pig. Sweet, Brit. Fl. Gard. t. 255 ; Gartenflora, t. 800 ; Loddiges, 
 Cabinet, t. 1506, sub. n. I. taurica. 
 
 Char act. Bulb ovoid, with fleshy, radish-like roots. Leaves 
 four to six in distichous tuft, falcate, so as to be nearly horizontal, 
 lanceolate, pointed, ribbed, glaucous, with a conspicuous horny margin. 
 Flowers two to three on a stem, so short that they appear sessile 
 among the young leaves. Flower two to three inches across. Tube 
 short, but rather longer than the inflated green spathe-valves. The 
 claw of the fall, which bears a toothed median ridge, high and con- 
 spicuous at the hind part of the blade, lower on the claw, is expanded 
 laterally into wings which are 
 transparent, and which embrace 
 the style. Standards minute, 
 spoon -shaped, spreading horizon- 
 tally. Crests of style large, qua- 
 drate. All parts of the flower a dull 
 greenish yellow, except the me- 
 dian ridge of the fall and its 
 neighbourhood, which is a brighter 
 yellow, marked more or less with 
 violet spots. 
 
 Var. MAJOR (TTJRKES- 
 TANIC A). Larger in all its 
 parts, in foliage and flower. 
 Flowers four or five on a distinct 
 stem, which is, however, wholly 
 hidden by the clasping bases of the 
 leaves. Blade of claw yellow, with 
 
 little or no tinge of green ; median ridge full yellow or orange, 
 toothed, or even fringed, with hair-like processes ; wings of claw very 
 Dale and transparent, almost colourless. In other features same as type. 
 
 FIG. 62. IBIS CAUCASICA MAJOR. 
 
78 
 
 VI 
 
 Var. KHARPUT. Stem a foot or more high, but wholly hidden 
 by the clasping bases of the eight or more distichously arranged leaves, 
 which, six inches or more in length, narrow more gradually towards the 
 point than in the type, and have a less conspicuous horny margin. 
 Flowers five or so, sessile, in the axils of the upper leaves, three inches 
 or so across, of a greenish yellow, except the median orange ridge of the 
 fall. The claw of the fall has hardly any lateral wings, and the blade 
 is large and oval. The standards are larger than in the type, bent 
 vertically downwards, with the edges sharply curled in so as to form a 
 channel. Crests of style large and quadrate. The whole aspect of 
 the plant is^very different from both the type and the var. major. 
 
 Time. March. 
 
 Hab. The Caucasus, Kurdistan, Armenia, Turkestan. The var. 
 major (turkestanica) is found in Turkestan ; the var. Kharput in 
 Armenia, near Kharput. 
 
 19. I. ORCHIOIDES. Carriers, Rev. Hort. 1880, p. 337, 
 fig. 68. (After Orchis, " the Orchid-like Iris.") 
 
 Lit. Baker, Bot. Mag. xlvi. (April 1890), t. 7111 ; Foster, Gard. 
 Chron. 1889, i. 588. 
 
 Syn. I. caucasica var. major (Maximo wicz, Act. Hort. Petrop. 
 
 p. 417 ; Bull. Acad. St. Petersb. x. (1880) p. 688). 
 
 Fig. Baker ; Carriere, as above. 
 
 Charact. Bulb very large, sometimes as large as a goose's egg ; 
 fleshy roots, much less developed than in 
 T. caucasica. Stem a foot and a half or 
 two feet, bare in the internodes between 
 the clasping leaves, which are longer, 
 narrower, more gradually pointed towards 
 the apex than in I. caucasica, of a more 
 glossy green, especially on the upper sur- 
 face, with the horny margin much less 
 marked. Flowers from two to three 
 inches across, of a rich, full yellow, 
 generally free from greenish tinge, with a 
 bright orange ridge, and variable dark 
 violet spots, blotches, or lines on the fall. 
 Wings of claw, so conspicuous in I. cau- 
 casica, almost absent. Spathe-valves 
 green, narrow, clasping the tube, not in- 
 flated as in I. caucasica. 
 
 Time. March or April, rather later than J". caucasica. 
 
 Var. I. caucasica var. cserulea (Regel, Descr. PI. Nov. ix. (1884) 
 p. 37) differs from above only in colour, which is a pale blue or 
 lavender, except that the ridge of the fall and its surroundings are 
 yellow, with lavender blotches ; sometimes the whole blade is a 
 creamy yellow. 
 
 St. 
 
 FIG. 53- IBIS OBCHIOIDES. 
 
79 
 
 J. caucasica var. oculata of Maximowicz (op. tit.) and Regel 
 (op. tit.), with yellow flowers, the fall being marked with blue blotches, 
 is probably a variety of this. 
 
 I. caucasica var. linifolia, Regel (op. tit), with yellow flowers and 
 linear leaves only 3 mm. broad, is also probably a variety of this. 
 
 Hab. Both the type and the above varieties are found in 
 Western Turkestan and in Bokhara. 
 
 NOTE. Having cultivated plants kindly sent to me by the late Dr. 
 v. Regel as J. caucasica c&rulea, I can speak with certainty as to these 
 having the characters of /. orchioides ; but not having grown the var. 
 oculata or linifolia, I cannot be sure whether these hold a like position 
 or no. Maximowicz seems (op. tit.) to state that even the type may 
 vary in colour and have a var. cxndea (this I have never seen) ; he 
 also insists that all the characters attributed to the several forms are 
 so mingled in different plants that all must be regarded as belonging 
 to one species. 
 
 20. I. PALESTINA. Baker, Journ, of Bot. 1871, p. 108, as 
 Xiphion Palsestinum. (From Palestine, habitat.) 
 
 Lit. Boiss. Fl. Orient, v. p. 122. 
 
 Charact. Bulb small. Leaves four, six, or eight, in a distichous 
 tuft, smaller and narrower than in J. 
 caucasica, striated on the outside, with a 
 finely toothed, horny margin, less con- 
 spicuous than in J. caucasica. Flowers 
 one to three, apparently sessile that is 
 to say, borne on a short stem, wholly 
 hidden among the leaves. Flower 1 to 
 2 inches across, like that of I. caucasica, 
 but with a much longer tube, narrow 
 (not swollen) spathe-valves, and a nar- 
 rower blade to the fall ; the minute 
 standards are toothed, and the rectangu- 
 lar crests of the style very large and con- 
 spicuous. Colour usually greenish yellow, 
 the blade of the fall being marked with 
 green, or with blue or violet veins, the 
 amount of blue being variable and some- 
 times prominent, as in the form named 
 var. cxrulea by Post. The median ridge of the fall, which is orange 
 in front, bears behind on the claw a number of black or deep violet 
 tubercles, each carrying a tuft of short black or dark purple hairs ; 
 a few longer dark hairs are scattered over the claw ; the wings of the 
 claw are not transparent. The flower is fragrant. 
 
 Time. December and January. 
 
 Hab. Palestine : Mount Hebron, Mount Carmel, Valley of the 
 Jordan, Sinai Peninsula. 
 
 FIG. 54.- I. PALESTINA. 
 
80 
 
 21. I. SINDJARENSIS. Boissier, Flor. Orient, v. p. 122. 
 (From Mount Sindjar, habitat.) 
 
 Pig. Bot. Mag. t. 7145. 
 
 Charact. Bulb very large, elongate, with fleshy roots. Stem 
 variable in height, sometimes a foot high. Leaves eight or ten, long, 
 and narrowing very gradually to a sharp point, bent into a double 
 channel, very striated on the outside, glossy green on the inside, 
 clasping by their bases the stem so as completely to hide it. Spathe- 
 valves narrow, pointed; as long as the tube. Flowers three or four, 
 two to four inches across. Fall with wings to the claw, a median 
 ridge, and somewhat narrow blade. Standards small, horizontal, 
 or turned down, spoon-shaped, with crenate edge. Crests of style 
 quadrangular, not very large. Blade of fall bluish white, with blue 
 veins. Median ridge and its surroundings yellow, with blue spots, and 
 beginnings of blue veins, which, more abundant on claw, give the latter 
 a blue colour. Standards and styles light blue. The whole flower is 
 blue in the centre, becoming paler at the periphery. Very distinctly 
 fragrant, odour resembling vanilla. 
 
 Time. March. 
 
 Hab. Mesopotamia, near the mountains Sindjar and Gebel 
 Taktak, and adjoining mountains. 
 
 22. I. FUMOSA. Boissier and Haussk. Sched. 1865 ; Boissier, 
 Flor. Orient, v. p. 123. (From famosus, smoky.) 
 
 Syn. Iris (Xiphion) Aucheri (Baker, Journ. of Bot. 1873, 
 p. 1010). 
 
 Charact. Differs from I. sindjarensis in having shorter and rela- 
 tively broader leaves, in the stem bearing a larger number, often eight 
 to ten flowers, and in the colour of the flowers, which are greenish 
 yellow, with more or less of a smoky tint ; in many respects resembles 
 I. orchioides or I. caucasica. 
 
 Time. April. 
 
 Hab. Syria, in neighbourhood of Aleppo. 
 
 23. I. STOCKSII. Baker, Gard. Chron. 1876, p. 723. (After 
 Dr. Stocks, Indian botanist.) 
 
 Lit. Boissier, Fl. Orient, v. p. 123. 
 
 Charact. Only known from dried plants. Leaves six or eight, 
 narrow, pointed, arched, with white margin. Stem short, bearing one 
 to three flowers, not unlike I. caucasica, but lilac or light purple, 
 though apparently varying in colour. 
 
 Hab. Afghanistan, near Quettah ; Beloochistan. 
 
 Time. Not flowered in this country ; probably winter-flowering. 
 
 NOTE. I cannot but think that if the various plants described from 
 dried specimens were brought into cultivation, it would be found that 
 they really represent more than one species. 
 
81 
 
 24. I. AITCHIS ONI. Baker, Journ. ofBot. xiii. (1875) p. 108, 
 as Xiphion Aitchisoni. (After Dr. Aitchison, Indian botanist.) 
 
 Lit. Boissier, Fl< Orient, v. 123. 
 
 Charact. Thin linear leaves, a foot or a foot and a half long. 
 Stem a foot or more high, bearing one to three flowers, and two or 
 three bracts. The claw of the fall does not bear wings. Flower 
 purple in the type, but in var. chrysantha bright yellow. By its 
 leaves, tall stem, and the absence of wings to the claw of the fall it 
 draws near to the xiphium group, but the fleshy roots, and the small 
 spreading standards, mark it as a Juno. 
 
 Time. Not cultivated. Flowers in its native home in March. 
 
 Hab. The Punjaub, near Otipore ; Afghanistan. 
 
 25. I. DREP AN OPHYLL A. Aitchison and Baker, Trans. 
 Linn. Soc. ser. 2, Bot. iii. 115. (From drepane (Grk.), sickle ; phyllon 
 (Grk.), leaf.) 
 
 Charact. Four to six narrow, arched leaves. A short stem, 
 bearing two, three, or more flowers, about two inches across. The fall 
 bright yellow, with the claw devoid of wings. 
 
 Time. Not in cultivation. Flowers in its native home in April. 
 
 Hab. Afghanistan, near Gulran. 
 
 26. I. ROSENBACHIANA. Regel, Descr. PI. Nov. pt. ix. 
 p. 35, tab. viii. (After General Rosenbach.) 
 
 Lit. Gartenflora, xxv. (1886) pp. 409, 617, t. 1227 ; Foster, Gard. 
 Chron. 1887, i. p. 90, and 1889, .p 
 
 i. p. 530 ; Baker, Bot. Mag. xlvi. 
 (Sept. 1890), tab. 7135. 
 
 Fig. Regel and Baker, as 
 above ; Garden, 1888, tab. 653, 
 tig. 2. 
 
 Charact. The bulb has 
 numerous fleshy roots, which 
 are, however, short and ovoid. 
 Leaves lanceolate, not very 
 pointed, eight inches or so long 
 and two broad, five or six to a 
 tuft. Flower sessile, appearing 
 soon after the leaves begin to 
 shoot ; one, two, or three to a 
 tuft. Tube of perianth from 
 two to six or more inches long. 
 Flower variable in size, but 
 often four or even more inches 
 across ; variable also in colour, 
 
 St 
 
 Fm - 5 ^. IRIS ROSENBACHIANA. 
 
 which is often very vivid, the prevailing hues being red- or blue- 
 purple, and yellow. Fall almost strap-shaped, the blade being not 
 
 F 2 
 
82 
 
 St 
 
 broader, but even narrower, than the claw, with a very conspicuous 
 median toothed ridge, which is generally a bright orange ; blade of fall 
 generally vividly coloured. Standard horizontal or drooping, narrow, 
 spoon-shaped, small, but relatively larger than in other Junos. Capsule 
 oval, with numerous reddish-brown, wrinkled, more or less cubical 
 seeds, each of which bears a conspicuous white strophiole. 
 
 Time. Flowers in March or April, sometimes in February. 
 
 Hab. Eastern Bokhara. 
 
 27. I. FOSTERIANA. Aitchison and Baker, Trans. Linn. 
 
 Soc. 2nd ser. Bot. iii. 114 ; Bot. Mag. 
 1892, t. 7215. (After Dr. Foster.) 
 
 Fig. Bot. Mag. t. 7215. 
 
 Charact. Slender, elongate 
 bulbs, with olive-green tunics ; fleshy 
 roots, thin and inconspicuous. Leaves 
 linear, not unlike those of I. xiphium, 
 but very striated on the outside. 
 A single flower, on a stem about a 
 foot high ; sometimes a second lateral 
 flower. Flower about two inches 
 across, with yellow falls and styles, but 
 red-purple standards. Claw of fall 
 narrow, not winged, suddenly expand- 
 ing into ovate blade. Standard hori- 
 zontal or deflexed, but relatively large. 
 
 Time. Flowers in March. 
 -Afghanistan, near Gulran. 
 
 28. I. ALATA. Poiret, Voyage en Barbaric, ii. 86. (From 
 alata, winged the winged claws of fall.) 
 
 Syn. I. scorpioides (Desfontaines, Fl. Atlant. i. p. 40, t. 6). 
 I. transtagana (Brotero, Fl. Lusit. i. p. 52). I. trialata (Brotero. 
 Phyt. Lusit. t. 95). I. microptera (Vahl, Enum. ii. 142). Xiphion 
 alatum (Baker, Jowrn. of Bot. ix. 1871, p. 108). Zhelysia arandifl-ora 
 (Salisbury, Trans. Hort. Soc. i. 303). Thelysia alata fparlatore, 
 Fl. Ital. iii. p. 317). Costia scorpioides (Wilkomm, Prodr. Fl. Hisp. 
 i. p. 144). Iris bulbosa latifolia of Clusius. 
 
 Fig. Bot. Register, t. 1876; Redout^, IM. t. 211; Gartenflora, 
 40 (1891), t. 1351 ; Bot. Mag. t. 6352 ; Brotero and Desfontaines, 
 as above. 
 
 Charact. Bulb ovoid, large, with several brown membranous 
 coats, and fleshy, finger-like roots. Leaves in distichous arrangement, 
 clasping and hiding the short stem, and bent into a channel, large 
 (up to a foot), lax, pointed, arched, pale green. Flower large, sessile, 
 generally jne, sometimes two. Spathe-valves lanceolate, somewhat 
 
83 
 
 swollen, pointed, more or less withered at flowering. Blade of fall 
 lanceolate, but claw expanded on each side in rounded, triangular 
 wings, which embrace the style. Standards very small, lanceolate, 
 spathulate, spreading horizontally. Crests of styles very large, 
 deltoid or quadrate, with toothed edge. Capsule ! oblong, buried 
 among the leaves. Seeds oval. 
 
 Hab. South of Spain, especially near Antiquera and Cordova ; 
 
 FIG. 57. IRIS ALATA. 
 
 South of Portugal, Monchique Mountains, Algiers, Morocco, Sardinia, 
 Sicily, Greece. 
 
 Var. C. Sprenger (B. Ort. Firenz. xiv. 1889, p. 13) describes 
 several varieties, differing chiefly at least in colour and size : var. 
 speciosa and lilacina from the Atlas Mountains ; nigrescens, cinerea, 
 and magna from Sicily, the last from Palermo. A white variety, 
 var. alba, is found in Spain and the Atlas Mountains. There is also 
 a var. cupreata, the origin of which is unknown. 
 
 29. I. NEPALENSIS. Don, Prodr. Fl. Nepal, p. 54. (From 
 Nepaul, habitat.) 
 
 Syn. 1. decora (Wallich, PI. Asiat. Ear. i. p. 77, t. 8G). See 
 also D. Don, Trans. Linn. Soc. xviii. p. 313. 
 
84 
 
 Pig. In Wallich as above, as I. decora ; Sweet, as I. nepalensis, in 
 Brit. Flow. Gard. 2nd ser. t. 11. 
 
 Charact. Rhizome small, covered and hidden by a nest of fibres, 
 into which the bases of the old leaves split up, bearing in the resting 
 stage a number of white, fleshy, thong-like roots. Leaves linear, ensi- 
 form, acuminate, striated, attaining their full growth after flowering. 
 Stem a foot or more, bearing a terminal bud of two to three flowers, 
 and sometimes a lateral bud, or even more than one. Spathe-valves 
 long, pointed, somewhat swollen, persistent. Tube of flower long, 
 
 rising above the spathe-valves, 
 and expanding in a funnel- 
 shaped perianth. Blade of 
 lanceolate-spathulate fall spread- 
 ing horizontally, bearing a 
 median ridge which extends 
 along the claw, and is fringed, 
 or even cut up into irregular 
 hairs. Blade of lanceolate- 
 spathulate standard spreading 
 horizontally, or even drooping, 
 so soon as the flower expands. 
 Crests of style large, deltoid or 
 quadrate, with serrated edge. 
 Stigmatic lip narrow, deeply 
 bilobed. Anther equal in length 
 to filament. Pollen coarse- 
 Fra. 58. IBIS NEPALENSIS. grained, white. Ovary, with 
 
 short peduncle, sharply trigonal. 
 
 Ripe capsule trigonal, with deeply grooved sides, dehiscing at apex. 
 Seeds oval, sharply pointed at one end, brown at narrow end, but 
 over the greater part of the surface covered by white strophiole. 
 Colour of fall at margin of blade light violet or lavender, in the more 
 median region marked with violet veins on a creamy- white ground, but 
 the median ridge is yellow or orange, with brown dots or blotches ; 
 standard lavender, with fine more deeply coloured veins \ styles. 
 lavender. 
 
 Hab. Mountains of Nepaul. 
 
 I. NEPALENSIS var. LETHA. Foster, Gard. Ghron. Oct. 
 1892, 458. (From Letha Mountains, habitat.) 
 
 Charact. Differs from above in being sessile. Fall with slight 
 constriction between claw and blade. Standard more distinctly 
 lanceolate. Ovary more rounded. Capsule shorter, broader. Seeds 
 more rounded. Foliage shorter, broader. 
 
 Hab. Upper Burmah. Letha Range of Chin Hills. 
 
 NOTE . Iris nepalensis is usually by Mr. Baker, for instance, and 
 
85 
 
 others included in the Evansia section or subgenus, by reason of the 
 conspicuous median ridge (crest) on the fall. As I have elsewhere 
 (Gard. Chron. 1887, i. p. 611) urged, though the group Evansia seems 
 to me a natural group, I cannot attach any importance to the ridge on 
 the fall. Indeed, in general I may say that the prominence, or want 
 of prominence, of a median ridge on the fall, and whether it is a mere 
 even ridge, or toothed, laciniated, or cut up into a beard, is about the 
 least trustworthy character that can be appealed to in determining the 
 affinities, and so the classification, of Irises. If the presence of a beard 
 is important as a token, J. Boissieri ought to go with I. pumila or 
 germanica in the Pogoniris group. If the distinction between a ridge 
 (crest) simply toothed and one made up of hairs (forming a beard) is 
 important, what are we to say to I. Hookeriana, in which the median 
 line is occupied behind by a dense beard, and in front by a toothed 
 crest 1 Indeed, something of the kind is seen often in Irises belonging 
 to the so-called Pogoniris section. I cannot place I. nepalensis in the 
 Evansia group simply on account of the character of the median 
 ridge of the fall, seeing, on the one hand, that this is almost the only 
 character by which it agrees with the other members of the group ; 
 and, on the other hand, the same character is almost as prominent 
 in some Juno Irises, such as J. caucasica turkestanica and J. alata. 
 I. / palensis stands almost alone. (I doubt whether it has any real 
 affinity to I. stolonifera of Regel, which bears similar fleshy roots, but 
 can say nothing definite, since I have not had this in cultivation.) As 
 I have urged in the text, J. nepalensis, by its almost bulbous nature, 
 by its fleshy roots, and by many features of the flower, comes nearer 
 to the Juno group than to any other, though it is markedly separated 
 from that group by the stem being sometimes branched, a character 
 which is absent from all Juno Irises. 
 
INDEX. 
 
 Synonyms are printed in Italics. 
 
 Costia scorpioides, 82 
 Hermodactylus bispathaceus, 74 
 longifolius, 74 
 repens, 74 
 tuberosus, 73 
 IBIS 
 
 cegyptia, 55 
 Aitchisoni, 40, 81 
 
 chrysantha, 40, 81 
 
 alata, 46, 48, 49, 52, 82 (Figs. 28, 57) 
 alba, 83 
 cinerea, 83 
 cupreata, 83 
 lilacina, 83 
 magna, 83 
 nigrescens, 83 
 speciosa, 83 
 Aucheri, 80 
 
 Bakeriana, 11, 60 (Figs. 9, 10, 38) 
 Battandieri, 23, 65 
 Boissieri, 24, 72 (Figs. 16, 17, 48) 
 Bornmulleri, 12, 62 
 bulbosa latifolia, 82 
 persica, 74 
 caucasica, 35, 49, 77 
 
 carulea, 38, 78 
 Kharput, 78 
 linifolia, 79 
 major, 36, 77 (Fig. 52) 
 major, 78 
 oculata, 79 
 turkestanica, 36, 77 
 DanfordiaB, 12, 61 (Figs. 11, 40) 
 decora, 50, 83 
 drepanophylla, 40, 81 
 filifolia, 22, 24, 27, 68 (Fig. 45) 
 Fontanesi, 23, 71 
 
 Fosteriana, 44, 82 (Figs. 26, 27, 56) 
 fugax, 55 
 fumosa, 40, 80 
 germanica, 50 
 Histrio, 7, 14, 57 (Fig. 6) 
 humilis, 59 
 imberbis, 72 
 juncea, 24, 71 (Figs. 18, 47) 
 
 numidica, 25 
 juncifolia, 56 
 Juno group, 31 
 Kolpakowskyana, 16, 63 (Figs. 12, 
 
 41) 
 
 Krelagei, 5, 6, 57 
 lusitanica, 20, 65 
 maricoides, 2, 55 
 mauritanica, 72 
 microptera, 82 
 monophylla, 2, 56 
 nepalensis, 49, 83 (Figs. 29, 30, 31, 58) 
 
 Letha, 51, 84 
 
 orchioides, 36, 78 (Figs. 22, 53) 
 caerulea, 38 
 
 IRIS 
 
 orchioides oculata, 38 
 
 linifolia, 38 
 
 Palestina, 37, 79 (Figs. 23, 54) 
 persica, 32, 49, 74 (Figs. 20, 21, 50) 
 Isaacsoni, 76 
 purpurea, 75 (Fig. 51) 
 pyrenaica, 68 
 
 reticulata, 4, 52, 56 (Figs. 3, 4, 5, 33) 
 cserulea, 60 
 cyanea, 6 
 
 group, cultivation of, 14 
 Histrio, 7, 14, 57 (Fig. 34) 
 histrioides, 9, 59 (Figs. 
 
 7, 37) 
 
 humilis, 59 (Fig. 36) 
 Krelagei, 5, 6, 57 
 major, 60 
 purpurea, 7, 60 
 sophenensis, 7, 58 (Fig.35) 
 Eosenbachiana, 41, 44, 49, 52, 81 
 
 (Figs. 25, 55) 
 samaritana, 56 
 scorpioides, 46, 82 
 serotina, 66 
 
 sindjarensis, 39, 80 (Fig. 24) 
 Sisyrinchium, 1, 55 (Figs. 1, 2, 32) 
 sophenensis, 7, 58 
 sordida, 20 
 spectabilis, 22, 65 
 Stocksii, 40, 80 
 tingitana, 23, 27, 70 (Fig. 46) 
 transtagana, 82 
 trialata, 82 
 tuberosa, 6, 28 (Figs. 19, 49) 
 
 cultivation of, 30 
 turkestanica, 36, 77 
 Vartani, 11, 14, 61 (Figs. 8, 39) 
 Winkleri, 17, 64 
 
 xiphioides, 18, 28, 67 (Figs. 14, 44) 
 Xiphium, 17, 20, 28 (Figs. 13, 42) 
 
 Battandieri, 23, 65. (Fig.43) 
 lusitanica, 20, 65- 
 "The Thunderbolt," 22 
 
 (Fig. 22) 
 
 group, cultivation of, 25 
 Mor&a Sicula, 56 
 
 Sisyrinchium, 56 
 Tenoreana, 56 
 Thelysia alata, 82 
 
 grandiflora, 82 
 Xiphion alatum, 82 
 Aucheri, 80 
 filifolium, 68 
 latifolium, 68 
 Palestimim, 79 
 persicum, 74 
 sordidum, 65 
 tingitanum, 70 
 vulgare, 64 
 
ADVERTISEMENTS. 
 
 , R, PEARSON & SONS, 
 
 BULBS! BULBS! BULBS! 
 
 The cheapest off er of First Class Bulbs ever 
 submitted to the Public. 
 
 DURING the past 10 years we have laboured to make the cultivation of Spring - 
 flowering Bulbs more general and more popular, and we are glad to say that our 
 efforts have not been without success. By offering to the Public the finest quality 
 of Bulbs obtainable, at the lowest profitable prices (for we do not wish to pose as 
 philanthropists doing business only for the pleasure resulting therefrom), we have 
 enormously increased our sales of these most charming plants. There are, however, 
 we find, still some prejudices to overcome in the mind of the Garden-loving Public 
 before we can carry this business to the limits which it ought to reach. 
 
 Firstly, many people write saying that our Bulbs are too cheap to be good, and 
 it takes time and experiment to convince them to the contrary. Again, we find that 
 many people send their orders to Dutch firms, under the impression that, by so 
 doing, they are dealing direct with the producer and obtaining the best value for 
 their money. To a certain extent this is true, but to how small an extent few are 
 aware; the fact is, one gets into the habit of talking and writing about early- 
 flowering Bulbs under the title of " Dutch Bulbs," without ever considering the 
 question of whether there is any justice in such a title, but a very little inquiry will 
 show how extremely small is the percentage grown in Holland. All the earliest 
 forcing Bulbs, as Roman Hyacinths, Paper White and Roman Narcissus, Freesias 
 and Lilium candidum, are grown in the South of France and Italy, and Lilium 
 Harrisii in the Bermudas. To order these from a Dutch firm is to pay carriage 
 from their place of growth into Holland and thence in small consignments to this 
 country. Then, again, take the Daffodils and Garden Narcissi, so deservedly 
 popular : we can produce far better and more healthy bulbs in this country than 
 elsewhere ; and who, with any knowledge of this subject, would order Snowdrops 
 from Holland whilst our home-grown roots were obtainable? We might pile 
 example upon example, but space fails. Suffice it to say that, without doubt, 
 Holland is the home, and ever will be, of the Hyacinth (the best growers do not, 
 however, export retail). Tulips also are grown very largely in Holland, and at 
 present command the market, but English Tulips are rapidly coming to the front. 
 Setting aside Tulips and Hyacinths, we could supply all other Bulbs usually 
 employed for early flowering if Holland closed her ports for ever. Under these 
 circumstances, we fail to see why general orders for Bulbs should be sent to 
 Holland instead of encouraging home industry. 
 
 OTHER SPECIALITIES. 
 
 Fruit Trees, Vines, Zonal Pelargoniums, 
 Hoses and Chrysantlieniunis. 
 
 IFI&IEIE TJIPOItT 
 
 ESTABLISHED 1782. 
 
 J. R. PEARSON & SONS, 
 
 Chilwell Nurseries, Notts. 
 
ADVERTISEMENTS. 
 
 ANT. ROOZEN & SON'S 
 
 BULBS, 
 
 oeou s Plants, 
 
 AND 
 
 IE IE ID S. 
 
 Our Descriptive CATALOGUES of the above, with full cultural 
 directions, are published in June and January, and will be sent, FREE, 
 on application to our Offices at OVERVEEN, near HAARLEM, HOLLAND, 
 or to our General Agents, 
 
 MESSRS. MERTENS & Co., 
 
 3 Cross Lane, St. Mary-at-Hill, LONDON, E.G. 
 
 ESTABLISHED 1832. 
 
 No Connection with any other Firm of the same Name. 
 
 THE GUILDFOBD HARDY PLANT NURSERY, 
 
 Established in 1888 for the culture especially of 
 
 ALPINE PLANTS AND HERBACEOUS OR OTHER 
 HARDY PERENNIALS. 
 
 Collections (or Selections) of Hardy Plants for any given soil, price, or purpose. 
 Special quotations for large quantities ; and dozens commonly at half the rates 
 charged for single plants. General Catalogue for 4 stamps (allowed to purchasers) 
 on application to the Manager, 
 
 GUILDFORD HARDY PLANT NURSERY, 
 MILLMEAD, GUILDFORD. l 
 
 GARDEN NETS AS FRUIT PROTECTOR^ 
 
 ABOUT OOO SQUARE YARDS AT 14s. EACH. 
 
 OTHER SIZES ACCORDING TO PRICE AND QUALITY. 
 
 Cut to order, id. and |d. per Square Yard. 
 
 IF O IR, IE ZHZ IF O IR, T T IR, .A. D IE S T7 IF IFL I E 3D. 
 
 DAVID PRATT, Net Merchant, 
 
 38 Rodger Street, 
 
ADVERTISEMENTS. 
 
 BARR & SON, KING ST., COVENT GARDEN, LONDON. 
 
 NURSERIES, LONG DITTON (SURBITON STATION). 
 
 Ready JANUARY. Sent free on application. 
 
 BARK'S Descriptive Catalogue of VEGETABLE SEEDS of the best 
 
 kinds only. 
 BARK'S Descriptive Catalogue of FLOWER SEEDS, including all 
 
 recent Novelties. 
 
 " Messrs. BARR & SON'S Seed Catalogue is a model of simplicity and clear arrangement. Those 
 interested in the production of the more uncommon sorts of Vegetables should certainly procure one of 
 these Catalogues. Messrs. BARR'S List of Vegetable Seeds is one of the most comprehensive in 
 circulation. Their List of Novelties in Flower Seeds is an interesting one, and many seeds which are 
 otherwise difficult to procure will be found in their general Flower Seed List." 
 
 THE QUKHX, January 22, 1802. 
 
 BEAUTIFUL HARDY 
 
 DAFFODILS 
 
 L/rll I UUILU 
 
 QOLD /V\EDAL /r 
 THE MOST LOVELY OF ALL SPRING 
 
 BARR'S HARDY DAFFODIL CATALOGUE contains full descriptions of 
 the Newest and Rarest Varieties. 
 
 SPRING EDITION ready in MARCH. 
 
 AUTUMN EDITION ready in AUGUST. 
 
 All Amateurs should possess a copy. Visitors are invited to inspect the grand display of upwards 
 of 2,000,000 Daffodils at BARB <fe Sox's New Nurseries, Long Ditton (Surbiton Station). 
 
 Ready SEPTEMBER. Sent free on application. 
 
 BARR'S Descriptive BULB CATALOGUE of Hyacinths, Tulips, Crocuses, 
 Bulbous Irises, Ranunculus, Anemones, Ixias, Amaryllis, &c. 
 
 " Messrs. BARR & SON submit to us their general Bnlb Catalogue, and their select list of Daffodils, 
 &c. These lists are illustrated only to a very limited extent, bnt they are practical as regards arrange- 
 ment and details. The little notes which accompany many of the articles betray the practised hand of 
 the writer, and the same appry to other features. The catalogues by no means make a beauty show on 
 paper, and rather suggest the sure way to secure a display of floral loveliness as the reward of the 
 intelligent cultivator." THE QUKEX, September 22, 1891. 
 
 Ready in SPRING. Sent free on application. 
 BARR'S Descriptive PLANT CATALOGUE of what is most effective 
 amongst HARDY HERBACEOUS PERENNIALS for Decoration in 
 Flower-Beds, Borders, Woodlands, and Wild Gardens, and of 
 what is most valuable for Rockeries and Rock-Gardens. 
 
ADVERTISEMENTS. 
 
 ESTABLISHED 1810. 
 
 SEEDSMAN, 
 C O RK, I RELA N D , 
 
 Besides Seed Trading to all parts of the world, WILLIAM BAYLOK HAKTLAND 
 
 has 
 
 10 ACRES of DAFFODIL GROUNDS at ARD-CAIRN, CORK. 
 
 The soil is a rich hazel loam, resting on limestone, and the bloom from the open this 
 
 year has been, from February 1, in the greatest possible profusion. All the Bulbs 
 
 go to rest end of April, and the lifting process begins in May with such early sorts 
 
 as Ard-Kigh, Golden Plover, &c. 
 
 "YEAR-BOOK" OF SEEDS FOR 1893, 
 
 IFOST IF:R,:E:E. 
 
 The CONFERENCE LIST of CULTIVATED DAFFODILS, an ALBUM, 
 
 or work of art not equalled in the world. Post free, 2s. Qd. 
 
 LITTLE BOOK: OF 
 
 in July. Post free. 
 
 WM. BAYLOR HARTLAND, F.R.H.S., 
 
 SEEDSMAN AND BULB GKOWER, 
 
 Telegrams: "Princepz, Cork." OA DATDIPIf CTRCCT PORK 
 
 Telephone direct with Grounds. 24 rAlKIUli OlilEEl, UUnlV. 
 
 KRELAGE'S BULBOUS IRISES, 
 
 j G-:R,:E:.A_T SDPZECI.A-IIIIT'X"- 
 TRUE TO NAME and OF THE BEST QUALITY ONLY. 
 
 For full particulars relating to BULBOUS IRISES (English and Spanish varieties, 
 and Asiatic species) and ONCOCYCLUS IRISES, please consult our Autumn Bulb 
 Catalogue. 
 
 For HERBACEOUS IRISES (germanica, pumila, Kcempferi, sibirica, &c.), see our 
 
 Spring Catalogue and special descriptive lists. 
 
 COLOURED PLATES of all species and varieties of Iris hitherto 
 introduced at 8d. and Is. each. 
 
 All our Catalogues may be had gratis and post-free on application to 
 E. BE. IKFgELAGE &, SO 1ST, a,a.yle (Moir^Md). 
 
 M ERRYWEATH E R'S 
 
 NEW GARDEN CATALOGUE 
 OF ENGINES 
 
 AND 
 
 WATERING APPARATUS 
 
 Contains many novelties which have 
 never yet been brought before the 
 Public. 
 
 63 LONG ACRE, LONDON, W.C.; & GREENWICH ROAD, S.E. 
 
ADVERTISEMENTS. 
 
 Estabd.] T. BXJUTO** &> 
 
 Horticultural Builders and Hot Water Engineers, 
 
 F.R.H.S., [25 Years. 
 BEXLEY HEATH, KENT. 
 
 All kinds of Horticultural Buildings erected on the most approved modern principles ; of 
 
 best material and workmanship. Personal Superintendence during erection. 
 
 Experienced Men sent to all parts. Heating and Ventilating guaranteed. 
 
 T. Burton, Sen., will consult icith Gentlemen in any part of the Country; Travelling Expenses 3rd Class only 
 
 charged. Estimates, Plans, and Catalogues Free. 
 
 Twelve Houses in Working Order, including Fruit Houses, Orchid Houses, Plant Houses, &c., at our Works. 
 
 Inspection invited to see improvements in construction. Charges strictly moderate. Testimonials from 
 
 Noblemen and Gentlemen from all parts of the Country. Architects' Designs carried out. 
 
 S. SHEPPERSON, 
 
 FLORIST AND SEEDSMAN, 
 PROSPECT HOUSE, BELPER, DERBYSHIRE. 
 
 SPECIALITIES Chrysanthemums, Carnations, Pinks, Pansies, Phlox, 
 
 Pentstemons, Pyrethrums, Fuchsias, Geraniums, Hardy Border 
 
 and Rockery Plants, &c., &c. 
 
 All warranted true to name, and the latest novelties up to date selected from the 
 best and most noted raisers. New List issued in January, gratis and post free. 
 
 All who have a Garden, Greenhouse, or Window should see my list before 
 
 elsewhere. Prices very moderate. 
 
 SEEDSMAN, 
 
 STIE^IEIET, HFSWIOHI, 
 
 Invites the attention of Amateurs to the 38th Edition of his 
 
 CATALOGUE OF FLOWTER SEEOS 
 
 Recently published. It includes many interesting novelties and rarities to be found in no other 
 Catalogue, in addition to all established favourites. No English Catalogue offers so large a collection of 
 
 HARDY BORDER AND ALPINE PLANTS. 
 
 As evidence of the reputation that his Seeds have obtained, he bags to state (that [he has been 
 honoured with extensive and unsolicited commands for the World's Fair at Chicago. 
 
 Sent post free on application. 
 
6 ADVERTISEMENTS. 
 
 Everyone interested in Horticulture should possess a 
 Complete Set of the 
 
 JOURNAL 
 
 OP THE 
 
 Dorticultural Society 
 
 which has been regularly published SINCE THE YEAR 1805, and therefore 
 contains not only a unique and unbroken record of the Plants, Flowers, 
 Fruits, and Vegetables which have been cultivated during a period of 
 almost 90 years, but also valuable Papers from the pens of the best 
 scientific and practical authorities on Horticulture. 
 
 Fellows of the Society and others desirous of completing their sets 
 may still obtain the following back numbers from the R.H.S. Office, 
 117 Victoria Street, S.W., only : 
 
 Vol. VI., 1880, pp. 168. Proceedings of the Society. To Fellows, 
 Is. ; non-Fellows, Is. 6d. 
 
 Vol. VII., part 1, 1886, pp. 155. Eeport of the Orchid Conference, 
 1885, with Papers by the late Professor Reichenbach, H. J. 
 Veitch, F.L.S., James O'Brien, H. N. Ridley, &c. To Fellows, 
 3s. 6d. ; non-Fellows, 6s. 
 
 Vol. VII., part 2, 1886, pp. 154. Primula Conference and Orchid 
 Conference at Liverpool. To Fellows, 3s. 6d. ; non-Fellows, 5s. 
 
 Vol. VIII, 1887, pp. 394. On the Effects of Frost on Vegetation, 
 by the Rev. Professor Henslow,M.A.,F.L.S.,F.G.S. To Fellows, 
 2s. ; non-Fellows, 3s. 6d. 
 
 Vol. IX., 1887, pp. 230. The National Pear Conference : Descrip- 
 tive Catalogue of varieties, &c. To Fellows, Is. 6d. ; non- 
 Fellows, 2s. Qd. 
 
 Vol. XI., parts 1 and 2, 1889, pp. 162. Hungarian Grapes, 
 Burmese Potatoes, Tomatoes, Hyacinths, Daffodils, Orchids, &c. 
 To Fellows, 2s. ; non-Fellows, 3s. 
 
 Vol. XI., part 3, 1889, pp. 278. Roses, Irises, Strawberries, Car- 
 nations, Peaches, Nectarines, Conifers, &c. To Fellows, 2s. Qd.\ 
 non-Fellows, 5s. 
 
 Vol. XII., part 1, 1890, pp. 289. Report of the Vegetable Con- 
 ference : Papers on Asparagus, Winter Salads, Peas, Potatoes, 
 &c. ; Report of the Chrysanthemum Conference, with Papers 
 read thereat. To Fellows, 2s. Qd. ; non-Fellows, 5s. 
 
ADVERTISEMENTS. 
 
 BACK NUMBERS OF JOURNAL-^o^mwed. 
 
 Vol. XII., part 2, 1890, pp. 232. Winter Gardening, Salads, Spring 
 Flowers, Report of the Daffodil Conference, &c., &c. To 
 Fellows, 2s. 6d. ; non-Fellows, 5s. 
 
 Vol. XIL, part 3, 1890, pp. 244. Flowering Trees and Shrubs, 
 Herbaceous Pseonies, Hardy Plants for Mixed Borders, Fruit 
 Evaporating, the Hollyhock, &c. ; Reports of the Carnation 
 Conference and of the Fern Conference. To Fellows, 2s. 6d. ; 
 non-Fellows, 5s. 
 
 Vol. XIII., part 2, 1891, pp. 232. Cyclamen, Snowdrops, Hybrid 
 Rhododendrons, Cape Bulbs, Lachenalias, &c., &c. To Fellows, 
 2s. 6d. ; non-Fellows, 5s. 
 
 Vol. XIII., part 3, 1891, pp. 280. Alpine Plants, Tea-scented 
 Roses, Hardy Summer-flowering Perennials, &c., &c. To 
 Fellows, 2s. 6d. ; non-Fellows, 5s. 
 
 Vol. XIV., 1892, pp. 588. Contains a full Report of and Proceed, 
 ings at the Conifer Conference, held at Chiswick, October 7 and 
 8, 1891, and all the Papers read thereat ; Synopsis of the order 
 Conifer, by Dr. Maxwell T. Masters ; Pinetum Danicum, by 
 Prof. Carl Hansen ; Statistics of Conifers in the British Islands, 
 by Mr. Malcolm Dunn; Diseases of Conifers; Value of Conifers 
 for Planting, &c., &c. To Fellows, 7s. 9d. ; non-Fellows, 15s. Qd. 
 
 Vol. XV., part 1, 1892, pp. 120. Contains a Report of the Con- 
 ference on Michaelmas Daisies and Perennial Sunflowers, &c. 
 To Fellows, 2s. 6d. ; non-Fellows, 5s. 
 
 Vol. XV., parts 2 and 3, pp. 400. Contains a Report of the Begonia 
 Conference, and all the Papers read thereat ; also the Conference 
 on Apricots and Plums, besides Papers on Winter Vegetables, 
 Cultivation of Melons, the Florist's Tulip, the History and 
 Culture of the Fuchsia, Root-pruning Fruit Trees, &c. To 
 Fellows, 3s. 9d. ; non-Fellows, 7s. 6d. 
 
 A List of the whole of the Plants, Flowers, Orchids, 
 Ferns, Fruits, and Vegetables certificated by the Society 
 from the year 1859 to December 1892. Price 2s. 6d. 
 
ADVERTISEMENTS. 
 
 ft;R.lb. tbe prince of Males. 
 
 PORTABLE 
 SUMMER 
 
 PRICE 
 
 Can be flxed by handy 
 
 Z 
 
 o 
 o 
 
 ui 
 
 RUSTIC 
 HOUSE. 
 
 800 
 
 man in half-an-hou. 
 
 O 
 
 LU CO 
 
 X co 
 
 Q 
 
 I 
 
 Q 
 
 LU 
 
 or 
 
 O 
 
 z 
 < 
 
 SIZE 6 feet diameter inside. 
 
 G. W. KLLEY, 
 
 81 DULWICH ROAD, HERNE HILL, S.E. 
 
 The Largest Manufacturer of Rustic Summer Souse 
 and Rustic Work in England. 
 
 A Large Selection of Summer Houses always in stock at price 
 
 ranging from 3. 10s. to 100. 
 
 RUSTIC SEATS, CHAIES, VASES, TABLES, ABCHES, &c., ALWAYS IN STOCK. 
 
 Rustic Bridges supplied to order. P i ans and Es ti mates f ree if requirec 
 
 Send for New Catalogue, 72 pp. (3 Stamps). 
 
 ^f^ ^^ ROAD, HERNE HILL 
 
 outside Herne Hill Station (L C. & D. Railway), where Goods can always be 
 seen. Ten Minutes by Train from City and West End 
 
' 
 
 
 BOOKS ON GARDENING 
 
 By \V. WU.I.IAMSUX and MALCOLM DUXX. 
 
 THE HORTICULTURAL EXHIBITORS' HAND- 
 
 BOOK. A Treatise on Cultivating, Exhibiting, and Judging Plants, Fl 
 Fruits, and Vegetables. By W.WILLIAMSON, Gardener; Revised by MAL- 
 COLM DUNN, Gardener to His Grace the Duke of Buccleuch and Queensberry. 
 Dalkeith Park. Crown 8vo. 3*. 6d. 
 
 "A took no one interested in horticulture can afford to be without. ... It is not only the most- 
 practical, but, in its divisions and details, tbe most comprehensive and complete work of its kind.' 
 
 HORTICULTURAL TIMES. 
 
 By DAVID THOMSON. 
 
 HANDY BOOK OF THE FLOWER-GARDEN. 
 
 With Engraved Plans of grouping in Beds and Borders. By DAVID THOMSON, 
 Gardener to His Grace the Duke of Buccleuch. K.G., at Drumlanrig. Fourth 
 Edition. Crown 8vo. 5*. 
 
 best book on the subject yet written or likely to be written for a long time to emm-. 1 - FILLD. 
 'The cultural directions, if carefully complied with, will enable the non- professional floriculturist to 
 trow plants a* well as any gardener.' GAKDEXEKS* CHRONICLE. 
 
 By DAVID THOMSON. 
 
 THE HANDY BOOK OF FRUIT-CULTURE 
 
 CNI'KR GLASS. Being a series of elaborate Practical Treatises on the 
 Cultivation and Forcing of Pines, Vines, Peaches, Figs, Melons, Strawberries, 
 and Cucumbers. With Engravings of Hothouses, Me. Second Edition. 
 Crown 8vo. with Engravings, ~s. >d. 
 
 ' No work of the kind of which we have any knowledge is at all to be compared with this. . 
 book is one that ought to be in the hands of erery gardener, of every amateur fruit-grower, Bud, indeed, 
 of everyone who desires to Inow the best that can be known en the subject.' Sc<n> MAX. 
 
 By WILLIAM THOMSON. 
 
 A PRACTICAL TREATISE ON THE CULTIVA- 
 
 TION OF THE GRAPE VINE. By WILLIAM THO.M.- ON, Tweed Vineyard, 
 Tenth Edition, Revised. 8vo. 5*. 
 
 ' \Ve cannot too strongly recommend Mr. Thomson'? tresiti.-e as a thoroughly practical ai 
 cultivation of the vine,* JOLKNAL v HOHTICULTUKE. 
 
 By F. W. BURPJDGi;. 
 
 DOMESTIC FLORICULTURE, WINDOW-GAR- 
 
 DENIN<i, AND FLORAL DECORATIONS. New Edition, Revised and 
 Enlarged. Crown 8vo. with upwards of 200 Illustrations, 7x. 6d. 
 'This book will meet the case of thousands who love flowers, and know not how to begin or, having 
 
 begun, know not how to go on in collecting and cultivating them. . . . It is a model of paiL 
 
 accuracy and good taste.* GARDENKB8' MAGAZINE. 
 
 P. W. UUKIUIMIE. 
 
 CULTIVATED PLANTS: Their Propagation and In.- 
 
 provement. Including Natural and Artificial Hybridisation ; Raisirg from 
 Seed, Cuttings, and Layers; Grafting and Budding. Crown 8vo. with 
 numerous Illustrations, 12*. (>//. 
 f This is a book not for the ordinary villa gardener, but for the professional gardener, and f< -r 
 
 amateurs who. by their interest in tbe pursuit, and the skill and patience they brinKto btar ' 
 
 rank in succtHS and authority with the professionals.' STANDARD. 
 
 WILLIAM BLACKPOOL) A: SONS, Edinburgh and London. 
 
7.\= 
 
 E. D 
 
 UNIVERSITY OF CALIFORNIA LIBRARY, 
 BERKELEY 
 
 THIS BOOK IS DUE ON THE LAST DATE 
 STAMPED BELOW 
 
 Books not returned on time are subject to a fine of 
 SOctSr volume after the third day overdue, increasing 
 to SI 00 per volume after the sixth day.. Books not in 
 demand may be renewed if application is made before 
 expiration of loan period. 
 
 MAR 4 1927 
 
 MAftM 1327 
 
 UiRARY Uf f 
 
 M? 
 
 6 1931 
 
 Junl 4'c P 
 
 ' I 
 Y L 
 
 Jun24 ' 40 
 
 LOA 
 
 l51976 
 
 507n-8,'26 
 
 SILYE1 
 
 ROYAL HORTICULTURAL SOCIETY 
 
 GOLD, SILVER-GILT, SILVER, AND BRONZE MEDALS, 
 PRIZES AND CERTIFICATES, 
 
 OTHER FIRST CLASS SHOWS.