and has recently received from him the varietal name “curticaudatus.” In fully mature specimens of Acanthodrilus rosæ the average number of post-clitellian segments is about 226. I have, however, met with two individuals, one having only 37, the other 49. In such individuals the posterior extremity of the body is not attenuated as in normal specimens; it is of the natural size of the body where it abruptly terminates, and is almost flat. Similar specimens of Acanthodrilus novæ-zealandiæ and Octochætus thomasi also occur. The segments composing the flat posterior extremity narrow within each other in a series of rings towards the anus. These deformities are also common in Lumbricus purpureus, Eisen, another introduced species. I possess specimens of the latter species and L. rubellus, with the clitellum, as the Rev. Mr. Friend observes, “right in the middle of the body.” At the present time it is important to call the attention of zoologists to these matters, seeing that they occur in widely-separated and widely-distributed genera. The careful study of such abnormal forms would, unquestionably, yield interesting and valuable results, morphologically.

Excepting some remarks by the Rev. Mr. Friend on abnormalities in two species of British earthworms, little attention has hitherto been given by naturalists to their great variability. Eisen has also shown that the clitellum in some species of European Lumbricus is occasionally moved one or more segments back or forward on the body of the worm. Both these peculiarities in structure occur occasionally in both introduced and native worms. Referring to Lumbricus rubellus, var. curticaudatus,^* the Rev. Mr. Friend says, “I have failed so far to determine whether the effect is due to soil, height above sea-level, want of proper food, or otherwise.” I cannot clearly conceive how any of these causes could produce a structurally imperfect worm. I think it is more probably due to some defect in the embryonic development of the worm, and not to the effect of any cause after its emergence from, the cocoon. Animals belonging to both lower and higher groups than earth worms are occasionally born without certain limbs, or other parts of their bodies, the result of causes of which little is known at present. The var. curticaudatus, Friend, occurs among many thousands of the species living together in the rubbish-heaps on the banks of the Ashburton River. The heaps consist of rotten straw, old bags, paper, tins, &c, thrown out from the shops. When rotten the worms collect into and breed in them in vast numbers. The largest and finest developed specimens of several exotic species are generally obtainable in similar heaps wherever they accumulate.

Earthworms vary greatly according to the soil and situation they inhabit. At the Rangitata Gorge Octochætus multi porus and Acanthodrilus novæ-zealandiæ attain their maximum-size. The site they inhabit is composed of a deep, rich, blackish loam, about 10in. in depth, resting on a substratum of yellow clay. The worms, when taken out of the soil, are of a reddish colour; whereas at Ashburton and other districts where I have obtained them, on the plains they are bluish-white, and occasionally of a yellowish tinge. At Ashburton they are smaller than individuals inhabiting moist land nearer the ranges. At one place where we dug out cocoons of O. multiporus we have occasionally obtained short, stout forms of that species with a largely-developed clitellum. Acanthodrilus novæ-zealandiæ, and the smaller species annectens, are very variable in size and colour in several districts where I have collected them. The known species of endemic Acanthodrili comprise a gradational series not known to exist in other genera in any single zoological region, while their prototype, Deinodrilus benhami, is also endemic. A. rosæ is the most constant form in colour and size I have met with in the endemic Acanthodrili.

Exotic earthworms, acclimatised in New Zealand, are extremely variable. Fully-matured individuals of Lumbricus terrestris are plentiful, ranging from 1in. to 6in. in length; the former occur in poor soil, the latter in decayed or decaying vegetable matter. I, however, need not particularise any species, for, although exotic earthworms are rapidly invading cultivated districts in hordes, they all attain their maximum size in heavily-manured gardens, and especially in accumulations of decayed vegetable matter^*. In the case of Microscolex modestus, Rosa, I think it probable that the extreme disproportion in size and variation in colour of individuals of the species may be the only cause for supposing them to belong to two distinct species. The variability of earthworms has been almost overlooked by authors when describing new species, more especially in the genus Lumbricus and the allied genera. The Rev. Mr. Friend's “Revised Synonymy of British Worms”^† should be specially acceptable to naturalists in unravelling much of the confused synonymy of British worms.

The great disparity in size of individuals of the same species may be due to several causes. Worms living in rubbish-heaps and accumulations of decayed matter not reduced to soil are better nourished and less liable to checks—such as dry, cold, or hot weather—than worms living in ordinary garden-soil. The same remark applies to native worms living in districts where the soil they inhabit is deeper, and where the rainfall is annually greater. The decayed vegetable matter on which introduced worms thrive well is much more nutritious and more easily assimilated during digestion than garden-soil. The finely-developed worms living in and subsisting on the decayed humus are generally paler than others inhabiting any garden-soil. I hope in the near future to ascertain the exact period of growth of several species of worms from the time of emergence from the cocoon to maturity.

When describing the mode of escape of the young of Octochætus multiporus from the cocoon^* I stated that the cocoons generally burst longitudinally. There are, however, some from which the worm escapes leaving a round hole at the end of the cocoon without rupturing the side. The cocoon-tissue is drawn out a little at the thicker end, which opens around the young worm's body while escaping. The head is protruded first, and the young worm slowly emerges. When the young worm is of large size the cocoon ruptures along the side and allows it to escape more easily. In a few instances we have observed them emerge in a doubled position. In the months of July and August we collected cocoons of Lumbricus purpureus, L. terrestris, L. rubellus, and Microscolex modestus. In the first-mentioned species the young emerged from the cocoon in three and four hours. The young of some of the three last named required from six to twenty-four hours. The cocoons of O. multiporus and A. rosæ are about equal in size. The recently-emerged worms are, however, distinguished by the latter becoming pale-brown in a few days.

Extinction of Endemic Earthworms.—When tussock land is first broken up great numbers of the larger species of worms (Acanthodrili) are torn to pieces or otherwise injured by the plough, and die from the effects. If the land is afterwards ploughed for a few years in succession very few full-grown worms escape destruction—in fact, wherever the land is well cultivated, whether in fields or gardens, the native earthworms disappear in a few years. So long as the native land remains in its naturally solid and rich condition alien worms avoid it, and are very rarely found in it. When once ploughed the natural freshness and loosening of the soil promotes a vigorous growth of introduced plants, which soon impover-

ishes and changes its chemical nature. The native worms are unable to survive the cultivation of the land, and rapidly disappear from soils where natural or artificial manures are used. It is, however, very different with the introduced species—the more manure and culture the land receives the more rapidly they increase and spread. As alien earthworms take possession of newly-broken-up land they rapidly change its chemical nature from the original condition. The change is disastrous to the indigenous worms, as they are unable to survive the changed or chemically altered nature of the soil. Like several other introduced animals, the enormous increase and dispersion of alien earthworms is phenomenal. The last mild, wet winter was very favourable to their increase. After every heavy rain-storm vast numbers were to be seen crawling about on the footpaths and streets, and in gardens and fields. From the 12th August to the night of the 5th September the weather continued very dry. In the evening of the latter date a heavy rain commenced to fall, and lasted until next morning. The storm caused prodigious numbers of worms to rise to the surface. Many of them were swept into the side-channels of the streets in Ashburton, and borne by them to the main sewer flowing along the east belt of the town. Owing to the gratings of several of the side-channels becoming choked, or through being unable to discharge the great rush of storm-water, the worms borne by the channels were left crawling about the grating in barrow-loads. This, nevertheless, represents a small number of the worms that are occasionally carried away by street-channels and other streams during and after heavy rains. They were nearly all Lumbricus terrestris, L. I observed a few only of Allolobophora subrubicunda, A. turgidus, A. rubellus, and A. fœtidus amongst them.

The numerous watercourses now traversing the Canterbury Plains from the hills to the sea are in many places favourable for the preservation of native earthworms. I have obtained very large specimens of Octochætus multiporus dug out of the sides of a water-race flowing through a paddock that had been cultivated for ten years. In many paddocks several yards wide of unbroken native land are left on both sides of the water-races. In these strips of land the native worms are favoured with moisture, and attain to a large size. While staying for three weeks at Mostyn House, near Springston, during July and August last, I was able to note the actions of earthworms in the neighbourhood of Lake Ellesmere. The soil, although of a light, free nature, is very fertile, and yields excellent crops of grain, grass, and other products of the farm. The eminence on which Mostyn House is built rises about 30ft. above the general level of the district, and forms an interesting relic of the sand-drift age in the neighbourhood of

Banks Peninsula. The dominant species of native worms inhabiting the district—at least, the slopes around the house—are Acanthodrilus rosæ A. novæ-zealandiæ, and an undescribed species of the same genus. The three species were obtained in a solid piece of land which probably had been formerly ploughed, but not for eight or nine years. In the lawns and gardens introduced worms are extremely rare. The fact is noteworthy, as the district is longer settled than others where they now occur in millions. The three species named above are, as I have formerly explained, able to lubricate their bodies externally with mucus. In the friable, sandy soil of the Springston district these worms would enrich and make the soil more coherent than other species ejecting less-adhesive mucus.^* Both A. rosæ and A. novæ-zealandiæ are very common in the limestone districts of Oamaru and Albury, where much of the land is extremely adhesive and very rich. In a large paddock in front of Mostyn House, now the residence of G. Jameson, Esq., a small slow - flowing creek winds irregularly through it. In the small bends of the creek where the plough had not touched I found some introduced worms. The creek is subject to small floods, which soon subside. On the night of the 30th July a heavy rainstorm commenced, lasting all night. Next morning the creek was in high flood, and spread out several feet from its bed on the gently-sloping grassy sides. As the water subsided during the day the worms ejected large castings composed of black soil. They were ejected on the surface of the slopes as far up as the flood-water had reached. They were the largest and most perfect castings ejected by a species of introduced Lumbricus that I have met with. The castings were ejected by Lumbricus terrestris, L., some of the specimens being exceptionally large and robust. I observed a few fresh castings in several parts of the open paddocks, but they were of small size. On the day after the storm I examined about a mile of the Springston-Leeston Road for traces of worms. I collected fourteen specimens of Allolobophora subrubicunda, Eisen, and observed over a hundred individuals of L. terrestris. A considerable number of fresh tracks were visible on the road where the worms had crawled over during the night. In an old plantation of Eucalyptus globosus I found two specimens of Lumbricus purpureus, Eisen, beneath an old cake of cow-manure. I also found three specimens of Allolobophora

turgidus, Eisen, under a rotten wet sack. Doubtless a careful search in the district would reveal other species. The occasional flooding of small grassy gullies like the one at Springston is favourable to earthworms. The saturation of the grassy slopes, and the deposition of fine silt on the surface during every freshet, renders the soil moist and viscous, consequently compelling the worms to eject all their castings on the surface. In the hot summer months these small grassy gullies on the plains become dry, excepting, in some of them, an occasional stagnant pool. As the summer advances the worms gradually retreat into the bottoms of the gullies, and live under the fine silt and beds of dead leaves deposited by the water in winter.

If we study the remarkable work of New Zealand earth-worms in former ages their rapid extinction at the present time in many districts is truly regrettable. The covering of the Canterbury Plains, and the sand-flats and dunes south-west of Banks Peninsula, with rich mould is worthy the attention of all interested in the work of earthworms. No less remarkable is their work on the West Coast plains, on the shingle-slopes and terraces and on the morainic accumulations in the valleys and upper parts of the plains on both sides of the Southern Alps. On the slopes of some of the higher alps, and in swampy flats often formed in depressions of the latter, where the alpine flora generally flourishes in luxuriance, the accumulation or layer of vegetable mould is frequently of considerable depth. The greater rainfall in these higher regions, especially during the summer and autumn months, would be favourable to the work of earth-worms at a period of the year when the surface of the mountains and upland swamps are free from snow. To what altitude worms exist in the New Zealand Alps I cannot at present say. I have found them under partially-embedded stones on the slopes of the mountains overlooking the Clyde branch glacier of the Rangitata, and at an altitude of about 4,000ft. on Mount Peel. They lie beneath the partially-embedded blocks of stone on the mountain-slopes and -spurs. The undersides of these large angular blocks are moist, and afford good shelter for worms at rest. The surface of the soil beneath the stones is frequently marked with a network of worms' tracks. The latter are moistened with their slimy secretions, and smoothened by the worms crawling along them. The soil on the higher slopes is coarse, and generally mixed with sharp stones. It is, however, more often moistened with rain and drizzling mists during summer and-autumn than the land on the lowlands. The only worms I have found in these high elevations were some large greenish specimens of Acanthodrilus novæ-zealandiæ. The larger form, Octochætus

multiporus, occurs in great numbers in a portion of an old river-bed near Ashburton covered with only a few inches of soil. Their actions in such situations are a perfect illustration of their mode of covering the shingle-flats with mould as they have done the plains in former ages, now elevated higher above the rivers. Although some eight or more species of native worms, ranging from 1in. to 16in, in length, inhabit the plains, the work of all was necessary to reduce the once great expanse of shingle, sand, and glacier-mud to the present mould-covered, fertile state. The periodical flooding of the rivers now traversing the Canterbury Plains is conducive to the formation of soil on the shingle-flats over which the flood-water spreads. The deposition of beds of unequal depths of fine silt explains the cause of the uneven depth of soil covering some parts of the plains. An examination of sections exposed on both sides of the rivers shows the same process or action of the rivers to have proceeded continuously since they commenced to cut their channels lower in the plains. During the retreat of the glaciers the volume of water in the rivers was immensely greater than now, while it flowed at a higher level, and spread out extensively over the plains. Occasionally smaller streams branched off, leaving small plateaux of shingle between them and the main stream. As the latter wore their channels deeper the beds of the smaller, streams were silted up during high floods, often forming extensive swamps. I have observed the Ashburton River deposit 2ft. and 3ft. of muddy silt during a single flood in these older side-channels now elevated above the main stream. In a few years a sparse vegetation grows in these silty or fine sandy places, while the earthworms gradually advance, and work them and the intervening shingly parts into areas of fertile land.

A broad shingly area is a formidable task for earth worms to cover with soil; yet such was accomplished by native earthworms on the now fertile Canterbury Plains. The late Sir Julius von Haast, the illustrious geologist and explorer of New Zealand, described the origin of the plains as follows: “With the exception of some morainic accumulations in the upper parts, and the drift-sands round Banks Peninsula, and the partial lacustrine deposits filling the former extension of Lake Ellesmere, the whole of the plains were formed by the deposits of huge rivers issuing from the frontal end of gigantic glaciers.” In the same article also occurs the following: “We must conclude that the Canterbury Plains were formed by the outlets of enormous glaciers, large torrents bringing down with them the morainic matter thrown in their course at the terminal face, raising their beds and shifting their channels at the same time so as to form fan-shaped fluviatile accumulations con-

sisting of shingle, sand, gravel, and glacier-mud.”^* Since the close of the glacier epoch in New Zealand the native worms have worked unremittingly “reclaiming the wilderness.” They have worked vast areas or plains composed of fluviatile deposits, as described by Haast, into fertile land on both sides of the Alps. In Canterbury the land is of the finest quality, while in Westland it supports magnificent forests only equalled in beauty and value in a few other parts of the world.

In addition to the aid they received in working some parts of the plains by the deposits of fine silt, they would, occasionally, be assisted in others by high winds filling the interstices of the shingle with fine sand. The work was unmistakably performed by the larger species of worms. They are able to swallow the larger particles of sand or small stones, which become gradually reduced by attrition in their gizzards and intestines, while the abundance of slimy secretion with which they lubricate their body externally would render the soil-forming particles more cohesive. In one place near the riverbed where we frequently visit to observe their habits and collect cocoons the surface-soil is about 3in. in depth. Beneath the latter the subsoil is simply a conglomerate of small stones and worms' castings voided in the interstices. The voided earth is deposited in the tracks of the worms, and varies in diameter from ¼in. to ½in. It contains numerous small stones, also of varying size, according to the size of the worms depositing them. Their cocoons occur promiscuously in the mixed shingle and castings, no attempt being made to construct burrows. By the same process some of the sloping shingly terraces are slowly, but gradually, covered with mould by the actions of the large worms. Slowly; but gradually, they have climbed the mountain-sides to considerable heights, covering them with mould in their upward advance. A glacial moraine is not an inviting field for the actions of worms, yet in New Zealand many have been, and are now being, successfully covered with soil by the constant work of these lowlyorganized animals. How the mucous secretions of the large worms may act chemically in reducing the coarse soil in the earlier stages of formation cannot be approximately ascertained until an analysis of each is known.

In the introduction to Beddard's paper, from which the following abstracts are taken, some interesting remarks occur on the comparative structure of certain New Zealand species:—

“The species of Acanthodrilidæ,” he adds, “found in New Zealand have been hitherto referred to three genera. Six years ago I described, under the name of Neodrilus mono-

cystis,^* an Acanthodrilid differing from, the typical forms (included, within the genus Acanthodrilus)by the presence of but a single pair of atria and spermatothecæ. This worm has been lately reinvestigated by Dr. Benham, ^† who has confirmed and extended my original account. I may add that I have within the last few months received some more examples of the same worm. There is, accordingly, no longer any doubt as to the characters of this species. It is not, as I at first thought it might be, a mere abnormality of such a species as Acantho-drilus dissimilis. Whether this earthworm should be really referred to a distinct genus is another matter. I prefer, however, to leave the question alone for the present. In any case, there can be no doubt whatever about the generic distinctness of Deinodrilus and the recently-described Plagio-chæta^‡, This latter may conceivably be identical with Hutton's Megascolex sylvestris;^§ at any rate, that species is stated and figured by Hutton to possess numerous setæ,. arranged in couples, which is the principal external character of Benham's Plagiochæta.

“The remaining Acanthodrilidæ have been all referred by me^‖ to the genus Acanthodrilus. This genus comprises altogether some forty species, of which nine are inhabitants of New Zealand.^¶ Forty species are not, perhaps, an unwieldy number for a single genus; it has, nevertheless, been divided into two genera—Acanthodrilus and Benhamia—by Michael-sen. Benhamia, it should be remarked, includes Benham's Trigaster. In distinguishing the two genera, Michaelsen has not considered the characters of the New Zealand Acanthodrilidæ. Benhamia is the name applied to those Acanthodrilidæ. with a ‘diffuse’ nephridial system—that is, in which the nephridia are not paired, but open on to the exterior, by numerous pores. Added to this character, Michaelsen originally called attention to the fact that the species with a diffuse nephridial system possess a pair of gizzards, or, as in Trigaster, three gizzards. Later he was led, by a consideration of the species Acanthodrilus schlegelii, to consider the definition of the genus, and to use, as part of the generic diagnosis, the

“There is another character to which comparatively little attention has been paid which may prove to distinguish the genera Benhamia and Acanthodrilus. In the two species of Benhamia described in the present paper, as well as in B. stuhlmanni, there are no setæ upon the eighteenth segment where the ventral pair should be; these setæ are also absent from the seventeenth and from the nineteenth segments, or, rather, they are there replaced by the penial setæ. In all the species of Acanthodrilus, on the other hand, which I have been able to examine, the ventral setæ are not missing from the eighteenth segment, though they are not present on the seventeenth and nineteenth segments, being there replaced by the penial setæ. This is also the case with the genus Octochætus; in two of the species of that genus, at any rate—-viz., O. multi-porus and O. antarcticus—the ventral pair of setæ of segment xviii. are present and quite normal; on segments xvii. and xix. the ventralmost seta is present and unmodified. Octochætus antarcticus is, furthermore, remarkable for the fact that the setæ of the seventeenth and nineteenth segments appear at first sight to be present and normal; as a matter of fact, the outer seta, of the ventral pair is replaced by the penial setæ, which occur exceptionally in this species. These setæ are very much smaller than in Acanthodrilus dissimilis, for example, and can hardly be seen until they are examined microscopically, but they are undoubtedly there. The ventral-most seta of the ventral pair is not absent from these segments as it usually is, but is quite recognisable. This species-is, therefore, in a less modified condition than is any other of the species of Octochætus, or Acanthodrilus, or Benhamia. In A. annectens and A. paludosus, described in the present paper, there is, as in Octochætus, a single seta ventrad of the atrial pores.

“In all the species of Benhamia which I have dissected, the calciferous glands are different from, those of Acanthodrilus or Octochætus. These glands are, in Benhamia, reniform pouches attached to the sides of the œsophagus; in Acanthodrilus and Octochætus these glands present the appearance of swellings upon the course of the œsophagus. Moreover, in Benhamia there always appear to be three pairs of calciferous-glands, which may, perhaps, prove to be always in segments xv., xvi., and xvii.; they have been for the most part described as in these segments, and it is possible that in those cases (e.g., B. buttikoferi, Horst) where they are stated to occupy the fourteenth, fifteenth, and sixteenth segments a mistake of one segment may have been made; anyhow, the three pairs seem to be characteristic, and nearly, if not quite, universal.

“The species Acanthodrilus novæ-zealandiæ, A. assimilis, A. rosæ, and A. smithi (to be described in the present paper), I

very much wider (astero-posteriorly), but still triannulate; so, too, are the segments which immediately follow, though much narrower. After the clitellum the segments continue to be, triannulate.

“The clitellum extends from segments xiii.-xix.; it is at first complete, extending right round the body; but on segments xvi.-xix. there is a ventral median area without any glandular modification.

“The atrial pores are upon very conspicuous papillæ; the two of each side of the body are connected by a longitudinal groove, which is not straight, but has a semicircular outline, the convexity being dorsal. The oviducal pores are just in front of the ventralmost seta. The setæ are rather distant from each other; a somewhat greater distance separates the two lateral setæ.

“I have not seen any dorsal pores.

“The pharynx occupies the first four segments of the body; the gizzard is very elongated, with parallel margins; it measures 65mm. in length; the gizzard occupies two complete segments, the fifth and the sixth. The œsophagus bears in segment xvii. the single pair of calciferous glands, which present the appearance of an oval swelling of the œsophagus itself. The intestine commences in the nineteenth segment.

“The septa of some of the anterior segments are, as is so constantly the case with earthworms, strengthened and bound together with thin muscular strips, which occasionally pass through one septum to reach another lying behind it. The first septum, which is thin and transparent, divides segments iv./v.; it is traversed by a large number of muscular threads which bind the pharynx to the pariatis. The next septum is also thin and delicate in texture; it is attached at the end of the first third of the gizzard; a good number of the threads which bind the pharynx to the pariatis pass through it. The following seven septa are thickened; the last of them, therefore, bounds the thirteenth segment anteriorly.

“The dorsal vessel is completely double; the two tubes of which it is composed retain their individuality when they pass through the intersegmental septa. The dorsal vessel is, however, at first a single tube; it is not until the seventh segment that it becomes double. In this segment commences the supra-intestinal vessel, which is large and very conspicuous. In segments x., xi., xii., xiii. are the four pairs of dilated hearts; in a few segments, anterior to the tenth, are more delicate peri-œsophageal vessels.

“There are, as in Octochætus multiporus, a pair of large nephridia lying close against (in front. of) the first septum; from each of these a slender duct was traced forwards which

opens, it may be inferred, into the buccal cavity; I did not, however, succeed in seeing the actual, orifice. In the rest of the body the nephridia are also constructed upon the plan which is characteristic of O. multiporus; the tufts appear to be massed chiefly around the setæ.

“Both testes and ovaries occupy the usual segments, but the gonads are attached to the posterior wall of their respective segments, as they are in Acanthodrilus annectens and in Octochætus multiporus.

“The racemose sperm-sacs are in xi., xii.

“The spermatothecæ are elongated pouches in viii. and ix.; they appear to have numerous minute diverticula crowded round the duct near its external opening.

“The atria do not extend beyond their proper segments. A number of strong muscular bands, such as occur in Octo-chætus multiporus, pass from the lateral to the ventral walls, of segments, and serve, no doubt, to extrude the papillæ already spoken of, on to which the atrial pores open. There are no penial setæ.

“This species is clearly most intimately related to O. multiporus; indeed, it is not, a little difficult to separate the two; the difficulty, too, is increased by the variability of the larger species. This difference of size is the most obvious difference; and it is, I think, a difference that must be allowed. The variability of O. multiporus unfortunately concerns those very organs upon which I had at first attempted to lay stress as distinguishing the two. In some individuals of Octochætus ultiporus the gizzard is limited to the sixth segment, the second septum lying just in front of it, attached, therefore, to the œsophagus; but in other specimens this septum is inserted on the gizzard itself, which thus occupies two segments, as in Octochætus thomasi. In two individuals the single pair of calciferous glands are in segment xviii.; but in others, as is the case with Octochætus thomasi, in the seventeenth. Another possible distinction between the small and the large species concerns the dorsal vessel in O. thomasi, as already mentioned, this vessel is single until the seventh segment. In a specimen of Octochætus multiporus the dorsal vessel was single until the commencement of the sixth segment only; in this segment it became double; in another the single dorsal vessel became double at the septum separating v./vi., but immediately after the two halves became fused, to again divide about the middle, of the segment. The shape of the gizzard and its relative strength in the two species does appear to differ; in the smaller species it is proportionately longer and narrower than in the large species.

“The next new species cannot by any possibility be confounded with the foregoing. I name it after Captain Hutton,

who has done so much in describing the fauna of New Zealand.”

Octochætus huttoni, Beddard. Pro. Zool. Soc. Lond., 1892, p. 674.

The first specimens sent to Beddard were discovered in the rich moist soil on the banks of Albury Creek, in South Canterbury. It is a larger species than O. thomasi, and very sluggish in habits. When taken out of the soil it becomes, extremely limp and soft, and does not appear to thrive in flower-pots. All the specimens I brought home with me died in a few weeks, although they were placed in fresh soil, and kept moist in a cool place. Following is Beddard's diagnosis of the species:-

“The examples sent to me were all of approximately the same size; an individual selected for accurate measurement was 130mm. in length by 7mm. in breadth at the clitellum, It consisted of 233 segments.

“The colour during life was pink, the clitellum being white; this colour is due to the fact that the species, like O. multiporus, had no pigment in the skin.

“The anterior segments are much annulated. After the clitellum there is also, though to a less extent, an annulation of the segments.

“The clitellum occupies segments xiii.-xix. (xx.).

“The atrial pores are borne upon a very prominent fold overhanging on each side the ventral surface, which in this-region appears, in consequence, as if hollowed out. The two pores of each side are connected by a longitudinal furrow.

“The prostomium is short and wide, and is not continued by grooves on to the buccal segment, which is marked by numerous furrows.

“The setæ are in pairs not closely approximated.

“The dorsal pores commence between xi. and xii.

“The gizzard measures 8mm. in length, and appears at first sight to occupy four or five segments; it really corresponds to segment v., which is increased in size at the expense of neighbouring segments for its reception. The calciferous glands are in xv. and xvi.; they have the appearance of being merely dilatations upon the course of the œsophagus, which is the case with the other species of this genus.

“The intestine begins in xviii., but the typhlosole (which is very prominent) does not begin before segment xx.; it ends at about 60 segments before the end of the body.

“The first septum is in front of the gizzard. After the gizzard are six stout septa, and, following these, are two which are rather more developed than the rest, but not so strong as those which precede them. The nephridia are

diffuse, and there is a particularly dense mass of tubes in the anterior, segments, which seem to represent the mucous gland of Octochætus multiporus^*

“The dorsal vessel is double, and there are three pairs of hearts in x.-xii.

“All the gonads are attached to the front wall of their segments.

“The spermatothecæ (in viii., ix.) have, a minute clump of diverticula, presenting the appearance of a solid body, about, the size of a pin's head.

“There are no penial setæ.

“This species is evidently perfectly distinct from the last; it is, however, clearly referable to the same genus, if this genus be admitted. The diffuse nephridia, double dorsal vessel, separate setæ, absence of penial setæ, and anterior position of gizzard cause it to resemble the three other species which I refer to the genus Octochætus. It differs from O. multiporus. and from O. thomasi in the following points:-

• “(1.) The prominence of the atrial pores as seen from the outside.

• “(2.) The position of the gizzard in segment v., and its limitation to this segment.

• “(3.) The presence of two pairs of calciferous glands in xv., xvi.; in this the present species resembles O. antarcticus.

• “(4.) The existence of only three pairs of hearts instead of four.

• “(5.) Six thickened septa instead of seven.

• “(6.) The attachment of the gonads to the front wall of their segments; in this character O. huttoni resemble O. antarcticus.

“The next species which I describe is referable to the genus Acanthodrilus (s.s.).”

Acanthodrilus smithi, Beddard. Pro. Zool. Soc. Lond., 1892, p. 675.

I discovered the first specimens of this species underneath large stones lying among the tussocks on the flat beneath the main limestone rock at Albury. It is the most active form of all the indigenous species of Acanthodrilus yet discovered. Like A. rosæ, its colour is rich brown, which changes only slightly in alcohol. In the living worm the segments are very clearly defined, and impart a neat effect to the animal when crawling over the ground. Its rapid motions are similar to those of Lumbricus rubellus, Hoffmeister, and Perichæta, intermedia, Beddard. While in the act of crawling the an-

terior half of the body is frequently off the ground, and placed down until the posterior half is drawn forward. These alternate movements, or modes of progression, are repeated in rapid succession. It is a very sensitive species, and if touched a little roughly will twist itself violently about, often leaping several inches along or off the ground. These remarks may enable collectors to identify the species when met with. Herewith I give an abstract of Beddard's description of the species:-

“The general appearance of the worm is very different from that of the other New Zealand Acanthodrilidæ which I have so far had the opportunity of studying—so different that it was unnecessary to dissect the species in order to ascertain its distinctness. All the New Zealand Acanthodrilidæ with the exception of the present species are either devoid of pigment in the skin, or, if pigment is present, the worms are of a brownish colour. Acanthodrilus smithi is (after preservation in alcohol) of a violet colour, the clitellum being whitish-yellow; the ventral surface of the body is the same colour as the clitellum.

“The worms are slender, measuring up to 75mm. in length, with a diameter of 3mm. at the widest part of the body. An individual of this size consisted of 114 segments.

“The prostomium completely divides the buccal segment, as it does in the New Zealand species, Acanthodrilus novæ-zealandiæ, &c.

“The setæ are paired, and the pairs are equidistant, the body being thus divided into four equal areas. There is no difference in size between the setæ of different segments.

“The clitellum occupies segments xiii.–xix.; it is saddle.-shaped, and is not developed ventrally beyond the outermost of the two ventral setæ.

“I could find no dorsal pores.

“The nephridiopores are very evident; they alternate in position from segment to segment, as is the case also with other species of Acanthodrilus from New Zealand. They sometimes lie in front of the dorsal, sometimes in front of the ventral setæ. When they open in front of the ventral they appear to be particularly related to the outer of the two setæ. When, on the other hand, they appear in front of the dorsal setæ the orifice is in front of the innermost of the two setæ which constitute the pair.

“The spermatothecal pores lie on the border-line, between segments vii./viii. and viii./ix. They lie in front of and to the outside of the ventral pair of setæ.

“The atrial pores are upon segments xvii. and xix.; their position corresponds exactly with that of the ventral pair of setæ, which are absent from these segments. The ventral are,

on the other hand, present upon the seventeenth segment; and the sperm-duct pores lie a little to the outside of and in front of these setæ; their position, therefore, corresponds more accurately to that of the spermatothecal pores than do those of the atria.

“The internal structure does not present any special points of interest, being, on the whole, very similar to that of the other New Zealand Acanthodrili.

“The nephridia are alternate in position, as in A. dis-similis; this peculiarity is confined, as regards the genus Acanthodrilus, to the New Zealand species.

“The dorsal blood-vessel is single; there are four pairs of dilated hearts, the last of which is in segment xiii.

“The alimentary canal is furnished with a rudimentary, gizzard, which needs a microscopical examination for its demonstration; such as it is, it lies in segment v. There are no conspicuous calciferous glands, but in segments xiv. and xv. the œsophagus becomes wider, and its lining membrane much folded and very vascular. This region evidently corresponds to the calciferous glands of other earthworms; all doubt upon the matter appears to be removed by the discovery of crystals exactly similar to those which occur in the calciferous glands of other Oligochæta. The vascularity of the œsophagus is not limited to these two segments; from the tenth segment onwards its walls are vascular, though not so folded as in the two segments xiv. and xv. The intestine commences on the eighteenth segment.

“The gonads occupy the usual position; the sperm-sacs are in segments ix., x., xi., xii. The atria are like those of other Acanthodrilidæ, and each is provided with a bundle of penial setæ. These setæ are recurved at the extreme end; the extremity has two delicate wing-like processes, which, when the seta is viewed from above, give to the end an oval contour; the tip of the seta in this aspect is seen to be bifid. The absence of any ornamentation upon the setæ appears to distinguish the New Zealand Acanthodrilidæ, with the exception of Octochætus antarcticus, where it is only very slight.

“The spermatothecæ are, as is nearly universally the case with the Acanthodrilidæ (Acanthodrilus (Diplocardia) communis is, so far as I am aware, the only exception), two pairs situated in segments viii. and ix. Each pouch has three small diverticula, one of which is constantly in front of the septum.”

Acanthodrilus paludosus, Beddard. Pro. Zool. Soc. Lond., 1892, p. 677.

Although the first specimens of this species—the smallest of the known native Acanthodrilidæ—were found in a marsh, I have found them in the fine, damp, sandy soil near the river-

bed, and under half-dried cakes of cow-manure. In habits it is extremely inactive, and when touched or handled curls up and remains motionless for a considerable time. In the living worm the natural colour in some specimens is pale-pink, while others are almost transparent, and show the dorsal blood-vessel through the clear body-wall. It is a very distinct and handsome little worm, and forms pretty spirit specimens. The following is an abstract of Beddard's paper describing the species:—

“This is a small and slender worm, but I have not preserved any accurate notes of its dimensions; it was about 1in. in length, and something like 1mm. in diameter. This species is a near ally of Acanthodrilus annectens, which I have already referred to as possibly worthy of generic separation from the Acanthodrilidæ with paired nephridia.

“The present species has the same arrangement of the setæ, which are not modified upon any of the segments of the body. I did not describe, in my account of Acanthodrilus annectens,^* the fact that only one of the two ventral setæ is missing on the segments which bear the atrial pores—i.e., xvii. and xix.; the apertures take the place of the missing outer seta of the ventral couple; on the eighteenth segment both setæ of the ventral pair are present; the pore itself lies to the outside of the pair. Acanthodrilus paludosus shows exactly the same arrangement, and both species therefore differ from Acanthodrilus smithi and from the other New Zealand species of Acanthodrilus in this matter: in them the ventral setæ are entirely absent from the seventeenth and nineteenth segments.

“The clitellum was not developed, though in other respects the worm appeared to be fully matured.

“The gizzard lies in segments v. and vi., but only one-fourth of the organ lies in the anterior segment. Calciferous glands are, as in Acanthodrilus annectens, totally absent; the intestine begins in the twentieth segment; some of the septa are thickened.

“There is a mucous gland, and the nephridia are paired.

“The gonads are normal in position; they are situated on the posterior face of their segments, as is the case with A. annectens. Opposite to them are the funnels of the ducts, which are like those of other species, and occupy the same segments. The sperm-ducts, however, agree with those of A. annectens to differ from those of most other earthworms, in running within the thickness of the body-wall; they retain their individuality until just before the external aperture.

“The atria have no peculiarities of structure; there are no