The

tracting the soluble arseniate with water, acidulating the solution with nitric acid, boiling to expel carbon dioxide, neutralizing carefully with ammonia (the reaction should be faintly alkaline rather than acid), and precipitating the arsenic in the cold with argentic nitrate as the brick-red salt Ag, As O4. latter is thrown on a filter, washed well, dissolved in nitric acid, and the silver determined by titration with ammonium or potassium sulpho-cyanate, whence the arsenic can readily be calculated. The results communicated showed very remarkable concordance, and apparently a high degree of accuracy. The exact degree of accuracy does not appear; since the percentage of arsenic in some of the substances tested was not determined gravimetrically, but assumed to be that required by theory. By this method, 0.1 gr. of enargite yielded 19.03 and 19.09% arsenic in successive trials. 0.05 gr. pure pronstite gave 15.08% arsenic, while 15.15% is theoretically required. An ore mixture gave respectively, 3.26, 3.30, 3.19, and 3.25% arsenic in different trials. A copper matte yielded 0.47 and 0.46% arsenic in successive determinations. Antimony, the presence of which in solution would vitiate the results of analysis, is almost entirely excluded by the use of sodium carbonate in the fusion. In a mixture of the enargite above tested with stibnite, 19.13% arsenic was found. No experiments were made to test the solvent action of the ammonium nitrate in the solution on the argentic arseniate. The advantages claimed for the method are the great ease and rapidity with which a determination can be made, and the high degree of accuracy attainable, fully sufficient, at least, for technical purposes. (Col. sc. soc.; meeting Feb. 5.)

AGRICULTURE.

[879

Action of peat on insoluble phosphates.— In an extensive series of experiments carried out at the Moor experiment-station in Bremen, Fleischer finds that certain peats exert a very considerable solvent action on phosphates. The first experiments were made in the laboratory by intimately mixing finely ground peat and phosphate, adding water, and allowing the mixture to stand, usually for three days. Peat from the lowland moors showed no solvent action; but that from highland moors (sphagnum peat) acted upon the phosphates in every case but two, dissolving from three or four to over fifty per cent of the phosphoric acid present, according to the nature of the phosphatic material. The materials used may be arranged in about the following order, the more soluble first: pure dicalcic phosphate, precipitated tricalcic phosphate, fine raw bone, steamed bone, commercial precipitated phosphates, bone-ash, crude Mejillones guano, Lahn phosphate. The action appears to be due to the presence of free humic acid, which decomposes the phosphates. In several cases the action went so far as to produce free phosphoric acid. Addition of potash-salts was found to increase the solvent action. These results are entirely in harmony with those that have been obtained in fieldexperiments on these soils. Almost invariably, insoluble phosphates have given better results than soluble ones, the reason evidently being, that, owing to the small absorptive power of peat, the soluble phosphates are soon washed out of the soil, while the insoluble phosphates yield up their phosphoric acid so slowly that the plants can utilize most or all of it. Experiments were also made in composting phosphates and peat. Here, also, phosphoric acid was dissolved, but not to so great an extent as in the laboratory experiments, where a much more intimate

mixture of the materials was possible. From 0.6 to 9.2 per cent of the total phosphoric acid was dissolved. Potash salts increased the solubility of the phosphates. A large proportion of the phosphoric acid .was rendered soluble in ammonium citrate; that is, brought into a condition similar to that of the socalled reverted phosphoric acid. In connection with these experiments, Kissling has studied the effect of the presence of various salts on the action of peat upon phosphates. Potassium sulphate increased the action decidedly, potassium chloride to a less degree, and sodium nitrate and kainit hardly at all. Gypsum and calcium chloride decreased the solvent action, and potassium carbonate destroyed it altogether, presumably by neutralizing the humic acid of the peat. The effect of the potassium sulphate was found to be almost exactly in proportion to the quantity used. Although the solvent action of peat, and of peat and potash salts, appears to be comparatively slight on the large scale, it is not without importance; since, in the soil, it may continue for a long time, and the products of the reaction may be continually removed by the movements of water in the soil and the action of vegetation. Fleischer found, that, after his mixtures of peat and phosphates were washed out, the action appeared to begin afresh ; and something very like this must occur in the soil. — (Landw. jahrb., xii. 129, 193.) [880

-

GEOLOGY.

H. P. A.

The Bow and Belly River districts, NorthWest territory. The rocks of the foot-hills and east of the mountains, according to G. M. Dawson, are entirely of cretaceous and Laramie age, overlain by bowlder clay and other beds referable to the glacial epoch. The geology of the region is complicated by the fact, that, in the immediate vicinity of the mountains the beds change considerably in lithological character, the change being such as would be expected to occur on the approach to a shore-line. So far, no reason has been found to suppose that any beds newer than the Laramie (including under this general name the Judith River and Fort Union series) have been found in this district, or, indeed, in any part of the Canadian North-West territory. The general arrangement of the rocks is given in the following table:

I. Laramie (including Judith River series).-1. Beds of the Porcupine Hills: massive sandstones, with shales, etc. 2. Willow Creek beds: reddish and purplish clays, with gray and yellowish sandstones. 3. St. Mary River series: sandstone shales and clays of general grayish or grayish-green colors. 4. Yellowish sandstones and shaly beds, with a mingling of fresh-water and brackish or marine mollusks.

II. Fore Hills.-1. Yellowish sandstones, with some shales, apparently irregular in thickness and character; mollusks all

marine.

III. Pierre group.-1. Blackish and lead-colored shales, with occasional sandstone intercalations, especially toward the mountains.

IV. Niobrara? - Belly River series: sandstones, shales, and sandy clays. Upper part generally grayish; lower, yellowish, and often banded by rapidly alternating beds. Fresh and brackish water mollusks.

Near its base, the Laramie of this region is a persistent lignite or coal-bearing formation. In the Pierre group, the most persistent coal-bearing horizon is at its base, although there is a coal-seam at its summit on Bow River. Mr. Dawson considers the coalbearing horizon at the base of the Pierre to be nearly equivalent to that at the base of the Chico group, which yields the coals of Vancouver Island at Nanaimo and Comox. (In this connection it is well to remember that the identity of the so-called Chico of Vancouver Island with the group of that name in

California is not by any means established.) The following approximate estimates of the quantity of coal underlying one square mile of country in several localities have been made:

Main seam, in vicinity of Coal Banks, Belly River, 5,500,000 tons.

Grassy Island, Bow River (continuation of Belly River, main seam), 5,000,000 tons.

Horse-shoe Bend, Bow River, 4,900,000 tons. Blackfoot Crossing, workable coal in seam as exposed on Bow River, 9,000,000 tons. (Geol. surv. Can.) J. B. M. [881 Triassic traps and sandstones. Mr. W. M. Davis last summer visited a number of localities in Massachusetts, Connecticut, and New Jersey, for the purpose of studying the relation of the trap masses to the triassic sandstones and shales. Some of these are dikes traversing the strata at high angles, and about such there has been comparatively little conflict of opinion. But the greater number exist as sheets conforming to the bedding; and these have been regarded by some writers as contemporaneous, by others as intrusive. Mr. Davis finds distinct evidence that some of the sheets were extravasated during the deposition of the strata, being afterward buried as the sedimentation progressed; and he finds equally distinct evidence that other sheets were injected between sedimentary layers already formed, and cooled under pressure. To the first class belong the principal masses of the Connecticut valley, including Deerfield Mountain, Mounts Tom and Holyoke, and the Hanging Hills near Meriden; to the second, belong the East and West Rocks near New Haven, and the Palisades of the Hudson. The principal intrusive masses occur in what are regarded as the lower portions of the formation, and may have been injected while the upper strata were still in process of formation.

A duplication of trap-ridges by faulting is demonstrated in some instances, and suspected in others; and it is pointed out that these faults may belong to a wide system, whose total effect is greatly to expand the outcrop of the formation by duplication. Each of the greater triassic districts presents a wide expanse of strata, with a prevailing dip at a considerable angle in one direction. To account for the phenomena by tilting alone, assumes an amount of deposition and subsequent erosion appalling even to the geologist; while the erosion demanded by the hypothesis of tilting and faulting combined is readily admissible.

The observations are prefaced by a bibliography of the subject, and followed by a general discussion, which includes an excellent digest of the opinions and observations of earlier writers. The paper makes a pamphlet of sixty octavo pages, illustrated by three plates. (Bull. mus. comp. zoöl., geol. ser., i., no. ix.) [882

G. K. G.

Ore-deposition by replacement.

As a result of his geological studies in Leadville, Col., Mr. S. F. Emmons has reached the conclusion that the ' carbonate deposits' of that locality were not formed by the filling of pre-existent cavities. They belong to a class of deposits for which he proposes the name metamorphic, and which are produced by a metasomatic interchange between exotic matter and original rock material. In Leadville the original rock is a dolomitic limestone, 150 to 200 feet thick; and the replacement has occurred either at or near its contact with an overlying sheet of porphyry. The introduced or vein material consists of silica and metallic minerals. These were brought in solution by percolating waters, having been previously dissolved from the

associated eruptive rocks. In places the whole bed of limestone has been replaced, but in general only a portion. The equivalent vein occupies less space than the limestone; but, allowing for this difference, the thickness of vein and the thickness of residual limestone are complementary.

Mr. Emmons regards the class of metamorphic deposits as an extensive one, including a large proportion of the so-called fissure-veins, both calcareous and siliceous, of the Rocky Mountain region. — (Phil. soc. Wash.; meeting April 7). [883

MINERALOGY.

Products of the alteration of corundum. — The following are the results of observations made by F. A. Genth:

Alteration into spinel. At the Charter mine, Madison County, N.C., corundum occurs crystallized, and in cleavage masses of a grayish or white color. In the cracks of the same it can be noticed that a change has taken place; and in many cases this extends through large masses, converting the corundum into a massive greenish-black spinel, rarely showing octahedral crystals. The same has a gravity of 3.751. Scales of prochlorite, into which the mineral finally passes, are often present. Analysis of the carefully selected material indicates that it has the composition of a spinel.

Alteration into zoisite. — At Towns County, Ga., pink crystals of corundum are found, surrounded by greenish-white cleavable zoisite.

Alteration into felspar and mica. - The author cites many occurrences in which cleavable masses of oligoclase and albite surround a core of undecomposed corundum, also where the corundum is surrounded by flat, cleavable mica (muscovite) or a delicate fibrous mica (damourite). Sometimes the mica and felspar occur together; and the nucleus of undecomposed corundum appears on its exterior very rough, as if it had been eaten into. Numerous analyses are given to prove the identity of the decomposition products.

Alteration into margarite. This occurs more seldom than the alteration into potash mica; and in some cases scales of the latter are interposed between the margarite, which usually is compact in its nature. Specimens showing this alteration are from Jackson and Iredell counties, N.C., and from Unionville and Aston township, Penn.

Alteration into fibrolite. Specimens from near Norwich, Conn., and Burke county, N.C., show radiated fibrolite surrounding crystals of unaltered corundum. It seems as if, in many cases, the fibrolite had undergone a subsequent change into mica.

Alteration into cyanite. At Iredell and Wilkes counties, N.C., bladed cyanite is found surrounding, and evidently resulting from, the alteration of corundum. From the latter locality the cyanite has partially undergone a change into micaceous minerals. — (Proc. Amer. phil. soc. Philad., xx. 381.) S. L. P. [884

GEOGRAPHY. (Arctic.)

Geographical notes from the north. The record of the Eira expedition appears in the Monthly record of geography for April, giving an account of the voyage up to Aug. 21, 1881, when the vessel was pierced by the ice, and the subsequent proceedings of the party until their rescue during the following summer. Even during the arctic winter, warm southerly gales occurred, which resulted in limited areas of open water. — Prof. Nordenskiöld's expedition will

ern coast.

sail some time during May, and will attempt a journey eastward over the ice from Auleitsivik fiord, in lat. 68° 30', near Egedesminde. Later an attempt may be made to penetrate northward along the south-eastNo new information has been obtained from the remainder of the Jeannette survivors, recently examined by the Naval board. In a recent lecture, Mr. E. H. Hall stated that the population of Newfoundland and Labrador amounts to 190,000, about one-quarter of whom subsist by the fisheries, which are valued at four and a half millions of dollars annually. The copper-mines produce about 45,000 tons of metal annually. A hurricane in British Columbia recently destroyed four vessels in Victoria harbor, and was attended with some loss of life. -The fur-seal fishery off Cape Flattery has been very productive this season, over 20,000 seals having been secured. The Newfoundland hair-seal fishery has also been remarkably successful, more than 200,000 hooded and harp seal being reported taken. On the other hand, the Dundee fleet, in the same waters, is said to have made a poor catch. -Ensign Stoney, U.S.N., will sail early in May in the revenue-cutter Corwin to distribute the presents from the government to the Chukchis, of St. Lawrence Bay, Bering Strait, who succored the crew of the U.S. S. Rodgers, which was burned in that bay while searching for the Jeannette party. -The growing scarcity of salmon for canning, in the Columbia River and southward, has led those interested to push into the undepleted waters northward. Several new fisheries have been established on the Skeena River, and others on the Chilkat River, and even in Cook's Inlet, nearly to latitude 60° N. - Four steam-whalers, built on the Pacific coast, will join the Bering Strait fleet this season. They are fitted with all the latest improvements, including iron tanks for oil and blubber, and are appropriately named the Orca, Bowhead, Narwhal, and Balaena. - It has been a very open season in Alaska, and in the south-eastern portion the snow was reported nearly gone March 25. aboriginal inhabitants of middle and northern Siberia, especially the Ostiaks and Samoyeds, are apparently either at a standstill, or even decreasing in numbers. According to recent investigations of Yadrintseff, their situation is precarious; and that they should gradually die out, as seems inevitable, is the more unfortunate, since many of them possess much intelligence and numerous good qualities. In Petermann's mittheilungen for April, Dr. Rink describes the investigations of the Danes in Greenland during recent years, in mineralogy, geology, geography, botany, and archeology, and gives a geological map of the west coast between Disco and Pröven. [885

W. H. D.

(South America.)

The

Chilian province Arauco.- A physical sketch of this province, by J. Sieveking, divides it into the littoral slope, the coast or Nahuelbuta range, the central plain, and the great Cordillera. The Nahuelbuta range extends north-north-west to south-south-east, and reaches an elevation of 5,000 feet. Its rocks are granite and crystalline schists, broken by basalt eruptions, and furnish gold to the streams. The auriferons gravels gave a rich yield to the early Spaniards, who forced the Indians to work them; but the latter rebelled, and drove away their would-be masters. In late years gold-washing has been again attempted with moderate success. Between the mountains and the coast is a hilly country, some twenty miles wide, rising to 1,000 feet elevation. It consists of Jurassic and later conglomerates and sandstones, which enclose valuable coal-beds, three to nine feet thick, with

a low percentage (two and a half) of ash. In Arauco little mining-work has yet been done: but, in the adjoining province on the north, the output reaches 10,000 tons a month; and with the rapid increase of steam-navigation along the western coast, of railroads in the interior, and of smelting and saltpetre works in the north, where fuel is scarce, this product must grow rapidly. The author believes the coal to be Jurassic, and not tertiary, as it has been described. All the coast range and littoral slope are heavily wooded, the climate rough and wet, especially in winter; the streams are short and not navigable, and, on nearing the coast, they cross a low plain of recent elevation. The harbors are open to the northwest, but closed on the south-west by the extension of sand-bars built up by the heavy waves and strong northward current. The central plain proves well adapted to agriculture and grazing at the few points where it has been settled; but the greater part is still unoccupied, except by the Araucanians, who maintain possession of a considerable share of good land in the south. Little is known of the Cordillera (it has hardly been entered), as winter begins early there with heavy snow-storms. The stones brought down by its streams are nearly all porphyritic, and sedimentary rocks are quite absent. There are two volcanoes on the range, Antuco, which Pöppig found active; and Villarica, near the lake of the same name. (Peterm. mitth., 1883, 57.) W. M. D. [886 (Africa.)

Lake Moeris. Another of the stories of Herodotus seems to be gaining ground. In 1871 RousseauBey found, by levelling, that the present lake, Birketel-Kerun, in the Fayum (cf. Schweinfurth, Zeitschr. f. erdk. Berlin, xv. 1880, 152, map), is at surface and bottom 41 and 55 met. respectively below the Mediterranean, and that its former level was 10 met. above the same datum, giving an original depth of 65 met. and a greatly extended area. Comparing this with the description of the Meridis lacus,' given by Herodotus, Mr. F. C. Whitehouse was confirmed in his trust of the old geographer, and, after some preliminary excursions, set out from Cairo early in 1882, and succeeded in finding by aneroid measurement a considerable depression south of Birket-el-Kerun, with its lowest point 180 feet below the Mediterranean, separated from the northern basin by a low divide (gisr), that seemed decidedly below the level of the Nile in this latitude. The southern end of this depression was not visited; but, as now mapped, the entire basin, if flooded from the Nile, might approach the area, and reach the depth, given. for it by Herodotus, although his description has generally been discredited, along with his assertion that it is manifestly artificial.' But this, also, Mr. Whitehouse seems to accept, as he speaks of the basin as avictory of mind over matter,' and suggests that we should treat the Mississippi as the Egyptians did the Nile. This conclusion, and the severely critical animus shown towards earlier writers, are the less satisfactory parts of the paper, which, in its evidence of work, its review of the cartography of the Fayum, and its quotations concerning Lake Moeris from ancient authors, contains much of interest. — (Bull. [887 Amer. geogr. soc., 1882, 85, map.) w. M. D.

Southern Abyssinia. - P. Soleillet writes from Ankober, Nov. 10, 1882, that he had made good progress, and obtained from King Menelik valuable concessions for the commercial company that he represents. A vast agricultural territory was open to their occupation and cultivation. Olive-forests were found to be very extensive: their fruit might be

Per

improved by grafting, and the company was conceded half of the yield for the next twenty-five years. mission was given to lay a narrow-gauge railroad from Obok, at the head of the Gulf of Aden, past lake Aussa, to Shoa, following up the left bank of the 'Ouache' (Hawash), where the construction would be easy and cheap. (Comptes rendus soc. geogr. Paris, 1883, 36.) (The road projected would be at least two hundred miles long, and partly in a very unproductive country; so that, in spite of the present activity of African development, this project can hardly expect an early completion.) — w. M. D. [888 (Atlantic Ocean.)

The Faraday Hills. — Dr. O. Krümmel has discussed the Atlantic soundings published by the Siemens Brothers (see 439), and shows that the Faraday Hills (about lat. 50° N., long. 30° W.) are very probably formed by submarine volcanic eruption. The soundings are so numerous and exact, that a trustworthy profile across the hills is constructed, exhibiting their surprisingly steep slopes (13 to 17° on one side, and 35 to 25° on the other), and revealing them as a mass about six miles broad at the base, rising from a bottom 1,300 to 1,700 fathoms deep to a summit about a mile broad in a minimum depth of 630 fathoms. Their form is therefore truly volcanic, and their altitude approaches six thousand feet. They are of rocky or stony surface, and have no ooze characteristic of deep-sea bottoms. The Flemish cape on the eastern slope of the Newfoundland banks is also stony, but this is regarded as a deposit of drift from melting icebergs. — (Ann. hydrog., 1883, 5, 146.) W. M. D.

[889

The Triton in the North Atlantic. - A sounding expedition on the British steamer Triton, under direction of Mr. Murray, formerly geologist on the Challenger, spent about a month in August and September last in exploring the Atlantic from the Shetland to the Faroe Islands, where, according to previous explorations, a shallowing of the bottom, named the Wyville-Thomson ridge,' separated the deep cold water on the north at 32° F. from the warmer bottom-water on the south at 47° F. In the northern part of the ridge, a depression was found with a depth a little over three hundred fathoms, through which some of the arctic water may pass southward. The shallower parts of the ridge, with a minimum depth of two hundred and sixty fathoms, is covered only with gravel and stones, and some of the latter showed distinct traces of glacial action. The fragments are of sandstone, diorite, mica-schist, gneiss, limestone, etc. Several new species were added to the faunae of the warm and cold areas first described in the results of the Lightning and Porcupine expeditions in 1868, 1869. — (Ann. hydrog., 1883, 194.)

W. M. D.

BOTANY. (Physiological.)

'890

Extravasation of water from leaves. This interesting phenomenon has been carefully examined by Volkens, who, while adding little that is really new, has shown the relations of the water-pores to the underlying tissues in a large number of families. It may be said, that, with three exceptions, the points of secretion were confined, in all cases examined, to the upper side of the leaf. The places are always distinguished by color, swelling, or some equally well-marked indication. The number of the pores is typical in many families and sections. (Jahrb. bot. gart. Berlin, 1883, 167.) G. L. G. (891

Continuity of protoplasm in contiguous cells. -Hillhouse's method is a modification of Sachs's, and consists in using dilute sulphuric acid on thin sections, following this by concentrated acid for several hours, thoroughly washing with water, and finally staining with ammonia-carmine. By this means it is possible to break down cell-wall without disturbing the protoplasmic threads. A similar process was used by Gardiner in his study of Mimosa. - (Bot. centralbl., xvi. 1883.) G. L. G. [892

Variable dichogamy. - Though as a general rule the Indian corn (Zea) appears to be imperfectly protandrous, beginning to shed its pollen one or more days before the stigmas of the same plant are developed, but continuing the process for several days afterward, in no small number of instances the dichogamy is reversed, so that the plant is strongly protogynous, while it is sometimes synacmic,-staminate and pistillate flowers maturing at the same time. This, with the similar behavior of the perfect flowers of certain species of Ribes, notably the golden currant, shows pretty clearly how either sort of dichogamy may have originated in what were at first synacmic species. (Rural New-Yorker, April 14.) [893

W. T.

Self-impotence of red clover. For six years Prof. Beal has been experimenting on the fruitfulness of Trifolium pratense, when self-pollinated and when crossed by humble-bees. Though the results obtained in the several years differ greatly, - from absolute self-sterility to the production of two-thirds as many seeds as by crossing, - they agree in showing a marked increase of seed where bees have worked. A source of error which tends to diminish the apparent value of crossing is the impossibility of excluding species of Thrips and other small insects by means of the netting used to cover the plants for the exclusion of bees, so that it is probable the degree of selfimpotence is greater than appears from these experiments. The general results may be gathered from the appended table of ratios:

First year. Second year Third year

Fourth year. Fifth year Sixth year

Average.

these may yet be found to occur in the West Indies, or on the mainland. One is Erigeron Darrellianus, with the habit and foliage of Conyza rivularis; the other, Statice Lefroyi, hitherto identified as S. Caroliniana. (Journ. bot., April, 1883.) s. w.

ZOOLOGY.

Mollusks.

[896

The position of Rhodope. The views of Graff (Morph. jahrb., viii. i.), referring Rhodope to the nudibranchiate mollusks, have received such wide publicity that it is well, even if a little late, that the views on this topic of the most eminent living student of the nudibranchs should have a hearing. R. Bergh of Copenhagen has examined Rhodope with special reference to the views of Graff, and finds, notwithstanding the fact that it is separated by marked characters from the ordinary turbellarians, that the differences between it and the nudibranchiate gastropod mollusks are much greater. There are no nudibranchs destitute of a heart, or of an organ filling the office of a kidney. Few have the liver reduced to a single mass. The genital organs of Rhodope do not differ greatly from those of turbellarians. The form and armature of the tail resemble those of many turbellarians, and nothing similar is known among the nudibranchs. Certain resemblances assumed to exist between the nervous system in Rhodope and Tethys, on the basis of Ihering's figure of the latter, have no force, since it appears that the figure is inaccurate. Lastly, a quietus is placed upon the theory by the fact that the larva of Rhodope has neither larval shell nor velum, which are universal in nudibranchs. It is therefore certain that Rhodope is no nudibranch, and eminently probable that it is nothing more than a peculiarly aberrant turbellarian. - (Zool. anz., 123.) W. H. D. [897

Fischer's Manuel de conchyliologie. — Part fifth of this excellent work is at hand, comprising pp. 417-512, which carry it forward from the Ascoceratidae, concluding the Cephalopods, through the Pteropods, and nearly through the order Pulmonata in the class of Gastropoda. The latter is divided as follows:

Subclass UNIVALVIA

Class GASTROPODA.

(Androgyna
Dioica

Order Pulmonata. Opisthobranchiata. Heteropoda. Nucleobranchiata. Platypoda.. Prosobranchiata. Subclass MULTIVALVIA . Polyplacophora. The author's paleontological researches have enabled him to preserve a satisfactory equilibrium as regards living and extinct forms. Numerous new and characteristic figures appear in the text, in addition to others not unfamiliar in the pages of Woodward; and with this fasciculus is added an atlas of twenty-four plates, which originally appeared in Woodward's Manual, and are well known, but which have never been excelled in clearness and accuracy by any purely black and white conchological plates issued up to the present time. The most casual inspection of the text, however, will show that we are presented with something quite different from a merely revised edition of Woodward, and that the volume when completed, though doubtless open to criticism in some of its details, will be by far the best text-book of the subject available. -W. H. D.

[898

Anatomy of Parmacella.-H. Simroth devotes a paper of forty-six pages, with an excellent plate, to the elucidation of the anatomy of P. Olivieri Cuvier. Its features are compared in detail with homologous organs in other pulmonates; and among his deductions the author concludes that the slugs constitute

calls attention to the fact that Fritz Müller did not keep the supposed Penaeus nauplius under observation until it changed to a protozoea, as is stated by W. K. Brooks in his recent account of the metamorphosis of Penaeus (Johns Hopk. univ. circ., Nov., 1882), and that, consequently, the rearing of the protozoea to the young Penaeus by Brooks proves nothing new in regard to the relation of Müller's nauplius to Penaeus. Faxon, however, sees no good ground for refusing to accept Müller's reasons for believing his nauplius and zoea stages to be parts of one lifehistory. (Amer. nat., May, 1883.) s. I. s. [901

Copepoda living in mollusks and ascidians. — C. W. S. Aurivillius has investigated the Copepoda inhabiting mollusks and ascidians on the Swedish coast, and published the results in two papers illustrated with seven double plates. Only two species, both belonging to the Sapphirinidae, were found inhabiting mollusks, - a species of Lichomolgus on species of Doris, and a new genus and species (Modiolicola insignis) upon the branchiae of Modiola and Mytilus. Twenty-one species, representing seven genera and five families, were found in the branchial sacs of ascidians, two new species being added to those already described by Thorell and others. Nearly all the old species are redescribed, and a large part of them figured, and analytical tables of the genera and species given. - (Ofvers. vet. akad. förh., 1882, Nos. 3 and 8.) S. I. S. [902

Insects.

Life-histories of American butterflies. - W. H. Edwards continues his careful and valuable descriptions of the early stages and habits of different American butterflies, giving us lately those of Grapta comma, G. interrogationis, and Pyrameis Atalanta. The descriptions of the caterpillars lose part of their value through lack of sufficiently explicit statement of the precise location of the dermal appendages.— (Can. ent., xiv. 189, 201, 229; xv. 14.) [903

Natural history of the fig-insects. — The very singular little group of fig-dwelling hymenoptera, referred by Westwood to the Chalcididae, is the subject of a recent monograph by Dr. Paul Mayer. Figgrowers have for ages taken advantage of the habits of Blastophaga grossorum for cross-fertilizing the tame fig with the wild caprificus. Mayer describes the anatomy of this species and some others, and discusses the geographical distribution of all known species, and their relations to the species of Ficus and its allies. The amount of adaptation induced by the peculiar habitat of the fig-insects varies in different