ALISMA RANUNCULOIDES VAR. ZOSTERIFOLIUM FRIES.

49

inflorescence showing no marked deviation from type, except in the presence of fascicled leaves on most of the panicles, which I believe to be analogous to the leaf-producing submerged rootlets. No fruit was seen, and I suspect that it seldom, if ever, occurs. Specimens from the River Laune, Killorglin, Co. Kerry, collected by Mr. R. W. Scully in August, 1890, and sent out under the name of var. repens, appear to me to be the same thing, though less well marked; an opinion endorsed by Mr. Scully himself and by Mr. Arthur Bennett, to whom I owe both the identification of my plant and almost all the information gleaned from books about it. He has looked through the material at Kew and South Kensington without finding anything similar to those above named, and, from an examination of Davies' specimens of his A. repens, concludes that the two forms should be kept separate. This appears to be rare, being only known hitherto from a few localities in Sweden, Denmark, Pomerania, and Holland. All the Floras that mention it treat it as a "good" variety; but whether it is really more than an extreme "state can only be proved by experiment. The first publication by Fries was in Botaniska Notiser for 1840, p. 35. In Nov. Fl. Suec. Mant. iii. p. 183, written two years later, ignoring his previous name, the author substituted that of sparganifolium, possibly considering it more appropriate. The earlier title must, of course, stand. He says:-" sparganifolium, foliis prælongis natantibus linearibus membranaceis. Bot. Not. 1840. In Elandiæ australis aquis G. M. Sjöstrand. Exacte respondet A. Plantagini graminifolio. Utriusque folia sunt phyllodia, in quorum apice laminam parvam abortivam videre licet." The original description runs:-" foliis longissimis linearibus natantibus (från Öland, Sjöstrand)." The following list (due to the source already mentioned) illustrates the book-history of the subject:

1753. Alisma ranunculoides L. Spec. Plant. ed. i. vol. p. 343. 1840. var. zosterifolium Fries in Bot. Not. p. 35.

1842. var. sparganifolium Fries, Mant. iii. p. 183.

1844. zosterifolium Fries in litt., Koch Synopsis Fl. Germ. et Helv.

1846. Fries Summa Veg. Scand. p. 65.

1864. y. littorellafolium Mortensen in Lange's Handb. i den Danske Flora, ed. 3, p. 795.

1868. Echinodorus ranunculoides G. Engelmann in Ascherson Flora d. Prov. Brandenburg, p. 651 (1864), var. foliis zosteraceis Buchenau. Abhandl. d. naturw. Vereines zu Bremen, xi. p. 17 (reprint).

1869. var. sparganifolium Fries. Marsson Flora von Neuvorpommern, p. 446.

1879. v. zosteræfolia Fr. i. Bot. Not. 1840, Hartman Skand. Flora, ed. xi. p. 416.

JOURNAL OF BOTANY.--VOL. 31. [FEB. 1893.]

E

50

AJUGA PYRAMIDALIS

IN SCOTLAND.

BY ARTHUR BENNETT, F.L.S.

In last year's Journal, p. 310, Mr. Colgan asks under what conditions, and at what elevations, the above species occurs in Scotland. As no one has replied to his query, I offer the following notes. In compiling them I am much indebted to Messrs. Miller and Duncan for notes on the species in the Hebrides and Sutherland.

Taking the counties in which it occurs, and in which the habitats are so stated as to be available :-In Orkney it occurs at about 600 ft., "on the sides of a hill." In the Outer Hebrides it grows among short grass about 100 ft. above sea-level on ground moderately dry. Another station is on the S.E. slope of one of the hills that occupy the peninsula at the S.W. corner of Harris, about 350 ft. above sea-level, on roughish, moderately dry ground, among short grass, and small tufts of heather.

In Caithness it grows on "The Old"; this is about 1250 ft. altitude, but I can find no note of the exact position of the plant on this hill. It also occurs on the grassy ledges of the cliffs on the north coast (about 300 ft.), and on the sloping banks (among grass) of one or two of the rivers at a low elevation (70′-100' ?). In E. Sutherland, on the sloping and rocky banks of a small burn near the coast; and again on the sides of the "Straths" on the east side of the watershed, bordering on the Caithness border. In W. Sutherland, about a mile inland, among rocks partly overshadowed by brushwood, about 2-300 ft. above sea-level; and in a grassy dell on the inland side of the sea-cliffs facing west, probably from 3-400 ft. altitude. In Dumfries, "on a small grassy plat formed by a slip in the rocky sides of the glen, at an elevation of about 1750 ft." (J. T. Johnstone) in the Moffat district.

Sir J. E. Smith describes its stations as "in dry pastures in the Highlands"; Hooker and Arnott as "Highland pastures." Mr. Bentham remarks (ed. 1), "It is never more marked than in recently burnt pastures"; this is the case in Sutherland, except that heather predominates over grass.

Looking beyond our own country, in Norway it extends upwards from 3500', 4000', and 5000'. Sommerfelt, in his Suppl. Fl. Lapponia gives "in graminosis humidis inferalpinum." Denmark, at a low elevation in the island of Bornholm, &c. (" in high grassy places"). In Belgium, in the glades of woods, pastures and heaths. In Italy, "in alpine pastures in the Alps."

In cultivation (from Sutherland) it often shows for flower in the end of March, and in early seasons is in full flower by the end of April, continning to the beginning of June.

Ajuga pyramidalis seems to be generally described as perennial, but it is often biennial, becoming perennial by buds in the lower axils of the leaves, which sometimes become very short stolons in the end of autumn. Last September Mr. Duncan sent me two

small plants that at the time puzzled me greatly as to what they could be, but growing on, they now show they are the Ajuga; the present leaves are curiously folded with patent hairs almost touching each other, looking much like a trap. Mr. Watson (Cyb. Brit, ii. 351) says: "Maintains itself by seeds in my garden in Surrey, but rather as a biennial than perennial."

I. THE BEST WAY TO MAKE MILLON'S REAGENT.

THE usual method of making Millon's reagent is that given by the text-books on physiology. The inconvenience in following the directions contained in those books is great, seeing that not only are nitrous fumes liberated in large quantity, much loss of time being caused before the fluid is ready, but the process is not feasible, supposing only a little of the reagent to be required. Seeing that Millon's fluid is well known as being a mixture of mercuric and mercurous nitrates, it would seem to be a matter for surprise if no attempt has been made to form a Millon's fluid by simply mixing the above nitrates in a certain proportion. As I have never heard of such an attempt, it may perhaps be worth mention that for some time I have used a Millon's reagent made by mixing the nitrates. Some preliminary experiments showed that a saturated solution of mercurous nitrate added to an equal quantity of mercuric nitrate as ordinarily sold, gives a fluid behaving in every way like one got by the action of hydric nitrate upon mercury. The advantages of this practice are that time is saved, there is no unpleasant smell caused, and just as much or as little of the reagent can be made-if it be only a few drops-as the operator requires.

II. A NEW WAY OF DEMONSTRATING CONTINUITY OF PROTOPLASM.

Within the last three years I have had much occasion to use Millon's fluid in connection with researches on callus and paracallus, and on the chemical constitution of cell-walls. Having frequently noticed that by careful boiling of sections mounted in Millon's fluid continuity of the slime through the sieves of sievetubes can often be made out in a beautiful manner, it occurred to me to try whether the fluid would be of any service in the demonstration of continuity through cell-walls in general. With ordinary tissues the result was not satisfactory, apparently because the boiling fluid acts too energetically upon the walls, but in the case of bony endosperms the reagent acts admirably if the precaution be taken of carefully applying heat to the preparation, when, in the course of a few seconds the intramural threads are well shown up. Preparations so treated may, after thorough washing, be mounted in glycerine, and they will keep for years. When it is

remembered that, except very rarely (e. g., Strychnos Ignatia), the ordinary methods employed to demonstrate continuity involve action of the reagent during several hours, the advantage of the plan here proposed is at once obvious.

III. ACTION OF COLD MILLON'S FLUID ON IRON-GREENING TANNIN, AND

ON CELL-WALLS GIVING PROTEID REACTIONS.

In a memoir recently published in Journ. Linn. Soc. vol. xxvii. I have endeavoured to show that the substance in certain cell-walls which causes them to give several of the reactions whereby proteids are recognised is not protein, as some continental authors (notably Weisner and Krasser) suppose, but is an iron-greening tannin. I was led to take up this position by the accumulation of evidence from several quarters; for not only did it appear that the cell-walls which will give some proteid reactions will not give others as distinctive, but the presence of iron-greening tannin could be demonstrated in those very walls. Moreover. it was found that solutions of iron-greening tannin behave exactly as do the walls to the various reagents employed, whether those reagents be reagents used in the detection of proteids, or reagents enabling us to discover tannin. Further, an attempt was made to explain why it is that certain cell-walls will take a distinctive colour with a given reagent, such as Schulze's solution, and some evidence was tendered in favour of the view that the presence of tannin (or at least of some glucoeide) often determines the colour taken in these cases.

As I am here writing about Millon's reagent, the opportunity is taken of stating that, in the course of some further researches on this interesting subject, an unsuspected confirmation of the above doctrine has lately come to light. I find that whereas when Millon's fluid is added to a solution of tannin, no change in the yellow ochre-coloured precipitate* ensues on allowing the unboiled product to stand overnight, yet that with an iron-greening tannin in the form of a solution of catechu, the result is quite different, since the precipitate slowly becomes brick-red without boiling. Here then is a crucial test which anyone who still favours continental views can easily apply. If the substance in the cell-walls which react like proteids be really protein, those walls should be unstained after lying overnight in cold Millon's fluid; on the other hand, staining of these walls would be evidence of a very decided character in support of the deduction advanced in my memoir.

The result of the experiments is here given in each case sections were kept overnight in Millon's fluid, but usually three or four hours' action is quite sufficient.

(a). Ivy. Xylem, hard bast and to a less degree outer cortical layers and epidermis stained as on boiling in the fluid. The stain also seen well in the sclerotised fundamental tissue lying upon the inner side of the xylem.

*If the solution be a strong one, the precipitate is at first orange, but it soon becomes ochre-coloured,

(b). Escallonia macrantha. Xylem and hard bast well stained. (c). Juncus conglomeratus. Xylem and sclerotised fundamental tissue surrounding vascular bundles well stained; walls of phloem less clearly stained.

(d). Yellow Jasmine. Walls of xylem, hard bast, phelloderm, and to a slighter degree of soft bast stained; sclerotic fibres running through cortex also well-stained.

(e). Privet. Xylem and hard bast stained.

(f). Pyrethrum Parthenium. Xylem and hard bast stained. (y). Berberis Darwinii. Xylem and hard bast stained.

(h). Maize. Walls of xylem and especially those of the sclerotised fundamental tissue in the neighbourhood of the vascular bundles stained.

(i). Rhizome of Arundo Phragmites.

Same as maize.

(j). Veronica sp. Hard bast and xylem stained. (k). Isoëtes lacustris. Meristem walls stained.

It must suffice to remark that these stained walls are precisely the walls which give the proteid reaction with boiling Millon's fluid. Moreover, iron-greening tannin in the cells of these plants, when it could be detected, reacted in the same way as did the walls to the cold fluid.

A PROVISIONAL LIST OF THE MARINE ALGE OF THE CAPE OF GOOD HOPE.

BY ETHEL S. BARTON.

I. PROTOPHYCEÆ.

LYNGBYA SEMIPLENA J. Ag. Sea Point, Boodle! A small specimen on Codium tomentosum.

CALOTHRIX CRUSTACEA J. Ag. Kalk Bay, Boodle!

Geogr. Distr. Adriatic.

DERMOCARPA PRASINA Born. On Rhizoclonium, Knysna, Boodle! On Cladophora rupestris, Cape, Harvey!

Geogr. Distr. North Sea. Adriatic.

II. CHLOROPHYCEÆ.

ULVEÆ.

ULVA LACTUCA L. Robben Island, Tyson! Kalk Bay, Boodle! Knysna, Krauss! Port Elizabeth, Sutherland! Port Natal, Krauss! No. 274; Gueinzius! Cape, Hohenack.! Meeralgen, No. 490; Reliquiæ Brebissoniana! Ser. 2, No. 206.

Var. RIGIDA. Kalk Bay, E. Young! Knysna, Boodle! Cape, Hb. Lenormand!

Geogr. Distr. N. Atlantic. North Sea, Mediterranean, West Indies.

U. FASCIATA Delile. Cape Point, Boodle! Kalk Bay, Boodle! Kei Mouth, Flanagan! Cape, Reliquia Brebissoniana! Ser. 2, No. 107. Geogr. Distr. General in warm seas.