2 Prof. Plateau’s chief communications will be found on pp.131 and 159; Mr. Nusbaum has furnished the account of the Development of the Cockroach, pp.180 to 195; and Mr. Scudder the Geological History of the Cockroach, chap.xi.

3 Correspondence of John Ray, p.142.

4 Copies dated 1762 have a plate representing the microscope and dissecting instruments used by the author.

5 Dufour. Rech. anat. et phys. sur les HÉmiptÈres (1833) les OrthoptÈres, les HymenoptÈres et les NeuroptÈres (1841), et les DiptÈres (1851). MÉm. de l’Institut, Tom. IV., VII., XI. Also many memoirs in Ann. des Sci. Nat.

6 Newport. Art. “Insecta,” in Cycl. of Anat. and Phys. (1839), besides many special memoirs in the Phil. and Linn. Trans.

7 Leydig. Vom Bau des Thierischen KÖrpers (1864), Tafeln zur vergl. Anatomie (1864), Untersuchungen zur Anat. und Histologie der Thiere (1883), &c., besides many special memoirs in MÜller’s Archiv., Zeits. f. wiss. Zool., Nova Acta, &c.

8 In some Insects there are traces of a fourth thoracic segment.

9 So also in some larvÆ (Calandra, Œstrus, &c.).

10 In some aquatic Insects the exchange of gases is effected by “pseudobranchiÆ,” and the tracheal system is closed.

11 Dragon-flies have the male copulatory apparatus, but not the genital aperture, in the fore part of the abdomen.

12 Aphis and Cecidomyia are at times viviparous, and a viviparous Moth has been observed by Fritz MÜller (Trans. Entom. Soc. Lond., 1883).

13 For descriptions of the species Fischer’s Orthoptera EuropÆa (1853) or Brunner von Wattenwyl’s Nouveau SystÈme des Blattaires (1865) may be consulted. The classification adopted by the last-named author is here summarised.

BlattariÆ.

A.—Femora spinous (SpinosÆ).

B.—Femora not spinous (MuticÆ).

Many useful references will be found in Scudder’s Catalogue of N. American Orthoptera, Smiths. Misc. Coll., viii. (1868).

14 LinnÆus was certainly mistaken in his remark (Syst. Nat., 12th ed.) that this species is native to America, and introduced to the East—“Habitat in America: hospitatur in Oriente.” He adds, “Hodie in RussiÆ adjacentibus regionibus frequens; incepit nuperis temporibus HolmiÆ, 1739, uti dudum in Finlandia.”

15 This must have been the “San Felipe,” a Spanish East Indiaman, taken in 1587. See Motley, United Netherlands, Vol.II., p.283.

16 Biblia NaturÆ, Vol.I., p.216.

17 De Borck. Skandinaviens rÄtvingade Insekters Nat. Hist., I., i., 35.

18 Brunner. N. Syst. d. Blattaires, p.234.

19 Scudder. Proc. Boston Soc. N.H., Vol.XIX., p.94.

20 For example, the Russians often call it Proussaki, the Prussian Cockroach, and believe that their troops brought it home with them after the Seven Years’ War. The native Russian name is Tarakan. In Finland and Sweden the same species is called Torraka, which appears to be a corruption of the Russian word, and confirms the account of LinnÆus quoted above.

B. germanica is found in the United States from the Atlantic to the Pacific. It is generally known as the Croton Bug, because in New York it is often met with about the water pipes, which are supplied from the Croton River (Dr. Scudder).

21 Bell’s Edition, Vol.I., p.454.

22 British Museum Catalogue of BlattariÆ (1868) and Supplement (1869). It is probable that the number is over-estimated in this catalogue, the same species being occasionally renamed.

23 Brongniart has just described a Carboniferous Insect which he considers a Thysanuran (Dasyleptus Lucasi), though it has but one anal appendage. See C.R. Soc. Ent., France, 1885.

24 Hummel, loc. cit.

25 The use of the term pupa to denote the last stage before the complete assumption of wings in the Cockroach, is liable to mislead. There is no resting-stage at all; wings are developed gradually, and are nearly as conspicuous in the last larval state as in the so-called pupa. There seems no reason for speaking of pupÆ in this case.

It is preferable to designate as “nymphs” young and active Insects, immature sexually, but with mouth-parts like those of the adult. See Lubbock, Linn. Trans., 1863, and Eaton, Linn. Trans., 1883.

26 See Appendix.

27 Insectorum Theatrum, p.138. The name Schwabe is frequent in Franconia, where it is believed to have taken origin. Suabia adjoins Franconia, to the south.

28 Compare the Swedish name (supra, p.18).

29 A fuller list of vernacular names is given by Rolland, Faune Populaire de la France, Vol.III., p.285. See also Nennich, Polyglotten Lexicon, Vol.I., p.620.

30 Anatomy of the Blow-fly, p.11.

31 Q.J. Micr. Sci., 1871, p.394.

32 Krukenberg. Vergl. Physiologische VortrÄge, p.200. Halliburton, Q.J. Micr. Sci., 1885, p.173.

33 Ann. d. Chem. u. Pharm., Bd. 98.

34 Previously observed by Leydig in Corethra.

35 A condensed and popular account of these researches will be found in Semper’s Animal Life, p.20.

36 Prof. Huxley (Anat. Invert. Animals, p.419) states that the integument splits along the abdomen also, but this is a mistake.

37 Audouin. Rech. anat. sur le thorax des Insectes, &c. (Ann. Sci. Nat., Tom I., p. 97. 1824.)

38 This application of the word to denote parts intermediate between terga and sterna has become general since its adoption by Audouin. It appears also in the older and deservedly obsolete nomenclature of Kirby and Spence. Professor Huxley has unfortunately disturbed the consistent use of this term by giving the name pleura to the free edges of the terga in Crustacea.

39 Where the thorax apparently consists of four somites, as in some Hymenoptera, Hemiptera, Coleoptera, and Lepidoptera, the first abdominal segment has become blended with it.

40 Balfour. Embryology, Vol.I., p.337.

41 E.g., by Graber. Insekten, Vol.II., p.423.

42 See, for example, Huxley on the Crayfish.

43 One of the few points in which we have to differ from the admirable description of the Cockroach given in Huxley’s Comparative Anatomy of Invertebrated Animals, relates to the articulation of the mandible, which is there said to be carried by the gena.

44 Morphologie des Tracheen-systems, p.103.

45 Zaddach, Entw. des Phryganiden Eies, p.86; Rolleston, Forms of Animal Life, p. 75, &c.

46 Zeits. f. wiss. Zool., Bd. XVI., pl.vii., fig.33.

47 Insekten, Vol.II., p.508.

48 Anat. Invert. Animals, p.398.

49 Professor Huxley has proposed to call the attached base hypopharynx, and the free tip lingua.

50 Professor J. Wood-Mason points out that in Machilis (one of the Thysanura) the mandible shows signs of segmentation, while the apical portion is deeply divided into an inner and an outer half. Ripe embryos of Panesthia (Blatta) javanica are said to exhibit folds which indicate the consolidation of the mandible out of separate joints, while the cutting and crushing portions of the edge are divided by a “sutural mark,” which may correspond to the line of junction of the divisions of a biramous appendage (Trans. Ent. Soc., 1879, pt.2, p.145).

51 The homology of the labium with the first pair of maxillÆ is in no other Insects so distinct as in the Orthoptera.

52 Rosenthal, Ueb. d. Geruchsinn der Insekten. Arch. f. Phys. Reil u. Autenrieth, Bd. X. (1811). Hauser, Zeits. f. wiss. Zool., Bd. XXXIV. (1880).

53 MÉm. Acad. Roy. de Belgique, Tom. XLI. (1874). Prof. Plateau’s writings will often be referred to in these pages. We owe to him the most important researches into the physiology of Invertebrates which have appeared for many years.

54 Exp. sur le RÔle des Palpes chez les Arthropodes MaxillÉs. Pt. I. Bull. Soc. Zool. de France, Tom. X. (1885).

55 Leydig, Taf. z. vergl. Anat., pl.x., fig.3. Hauser, Zeits. f. wiss. Zool., Bd. XXXIV., p.386. Jobert has figured the sensory organs of the maxillary palps of the Mole-cricket (Ann. Sci. Nat., 1872), and Forel similar organs in Ants (Bull. Soc. Vaudoise, 1885).

56 The reader who desires to follow this subject further is recommended to study chap. vi. of Graber’s Insekten, which we have found very useful.

57 Freshwater Crustacea, however, are sometimes similar to their parents at the time of hatching.

58 In Dytiscus the mandibles are perforate at the base, and not at the tip. See Burgess in Proc. Bost. Soc. Nat. Hist., Vol.XXI., p.223.

59 Ein KÄfer mit SchmetterlingsrÜssel, Kosmos, Bd. VI. We take this reference from Hermann MÜller’s Fertilisation of Flowers.

60 An interesting account of the structure and mode of action of the Bee’s tongue is to be found in Hermann MÜller’s Fertilisation of Flowers, where also the evolution of the parts is traced through a series of graduated types.

61 See Newport’s figure of Vanessa atalanta (Todd’s Cyc., Art. Insecta), or Burgess on the Anatomy of the Milk-weed Butterfly, in Anniversary Mem. of Boston Soc. Nat. Hist., pl.ii., figs. 8–10 (1880).

62 Balfour, Embryology, Vol.I., p.337.

63 Huxley, Med. Times and Gazette, 1856–7; Linn. Trans., Vol.XXII., p.221, and pl. 38 (1858).

64 “I think it is probable that these cervical sclerites represent the hindermost of the cephalic somites, while the band with which the maxillÆ are united, and the genÆ, are all that is left of the sides and roof of the first maxillary and the mandibular somites.”—Huxley, Anat. Invert. Animals, p.403.

65 Balfour, Embryology, Vol.I., note to p.337.

66 J.S. Kingsley in Q.J. Micr. Sci. (1885), has reviewed the homology of Insect, Arachnid, and Crustacean appendages, and comes to conclusions very different from those hitherto accepted. He classifies the appendages as pre-oral (Insect-antennÆ) and post-oral, and makes the following comparisons:—

Pelseneer (Q.J. Micr. Sci., 1885), concludes that both pairs of antennÆ are post-oral in Apus, and probably in all other Crustacea.

67 Many Orthoptera, which seize their prey with the fore legs, have a very long pronotum.

68 Also in PhasmidÆ (see Scudder, Psyche, Vol.I., p.137).

69 Professor Huxley (Anat. Invert. Animals, p.404) points out that the so-called pulvillus ought to be counted as a sixth joint. The same is true of the foot of Diptera and Hymenoptera, where there are six tarsal joints, the last carrying the claws. (Tuffen West on the Foot of the Fly. Linn. Trans., Vol.XXIII.)

70 The nomenclature adopted by Packard (Third Report of U.S. Entomological Commission) seems to us open to theoretical objections.

71 On wing-plaiting and wing-folding in BlattariÆ see Saussure, Etudes sur l’aile des OrthoptÈres. Ann. Sci. Nat., SÉr. 5^e (Zool.), Tom. X.

72 GrundzÜge der Vergl. Anat. (Arthropoden, Athmungsorgane.)

73 Origin and Metamorphoses of Insects, p.73.

74 PalmÉn cites one striking proof of the low position of EphemeridÆ among Insects. Their reproductive outlets are paired and separate, as in Worms and Crustacea.

75 These examples are cited by PalmÉn.

76 Eaton, Trans. Ent. Soc., 1868, p.281; VayssiÈre, Ann. Sci. Nat., Zool., 1882, p. 91.

77 Zur Morphologie des Tracheensystems (1877).

78 We take these instances from Eaton, Monograph of EphemeridÆ, Linn. Trans., 1883, p.15.

79 Charles Brongniart has lately described a fossil Insect from the Coal Measures of Commentry, which he names Corydaloides Scudderi, and refers to the Pseudo-Neuroptera. In this Insect every ring of the abdomen carries laminÆ, upon which the ramified tracheÆ can still be made out by the naked eye. Stigmata co-existed with these tracheal gills. (Bull. Soc. Sci. Nat. de Rouen, 1885.)

Some Crustacea are furnished with respiratory leaflets, curiously like those of Tracheates, with which, however, they have no genetic connection. In Isopod Crustacea the exopodites of the anterior abdominal segments often form opercula, which protect the succeeding limbs. In the terrestrial Isopods, Porcellio and Armadillo, these opercula contain ramified air-tubes, which open externally, and much resemble tracheÆ. The anterior abdominal appendages of Tylus are provided with air-chambers, each lodging brush-like bundles of air-tubes, which open to the outer air. LamellÆ, projecting inwards from the sides of the abdominal segments, incompletely cover in the hinder part of the ventral surface of the abdomen, and protect the modified appendages. (Milne Edwards, Hist. Nat. des CrustacÉs, Vol.III.)

80 Gerstaecker has found in the two first abdominal segments of Corydia carunculigera (BlattariÆ) pleural appendages, which are hollow and capable of protrusion. They have no relation to the stigmata, which are present in the same segments, and their function is quite unknown. See Arch. f. Naturg., 1861, p.107.

81 Jointed cerci are commonly found in Orthoptera (including Pseudo-Neuroptera); in the Earwig they become modified and form the forceps. The “caudal filaments” of Apus are curiously like cerci.

The cerci are concealed in the American Cryptocercus, Scudd. (Fam. PanesthidÆ).

82 Entw. der Biene. Zeits. f. wiss. Zool. Bd. XX. Or, see Balfour’s Embryology, Vol.I., p.338.

83 From more recent observations it is probable that abdominal appendages are usually present in the embryos of Orthoptera, Coleoptera, Lepidoptera, and possibly Hymenoptera. The subject is rapidly advancing, and more will be known very shortly.

84 See, for example, Klein’s Elements of Histology, chap. ix.

85 The exceptions relate chiefly to the alary muscles of the pericardial septum. Lowne (Blow-fly, p.5, and pl.v.) states that some of the thoracic muscles of that Insect are not striated.

86 For example, Prof. Huxley, in his Anatomy of Invertebrated Animals (p.254), says that “as the hard skeleton [of Arthropods] is hollow, and the muscles are inside it, it follows that the body, or a limb, is bent towards that side of its axis, which is opposite to that on which a contracting muscle is situated.” The flexor muscles of the tail of the Crayfish, which, according to the above rule, should be extensors, the muscles of the mandibles of an Insect, and the flexors and extensors of Crustacean pincers are among the many conspicuous exceptions to this rule.

87 Haller. This and other examples are taken from Rennie’s Insect Transformations.

88 Bull. Acad. Roy. de Belgique, 2^e. SÉr., Tom. xx. (1865), and Tom. xxii. (1866).

89 Loc. cit. 3^e. SÉr., Tom. vii. (1884). Authorities for the various estimates are cited in the original memoir.

90 Klassen und Ordnungen des Thierreichs, Bd. V., pp.61–2.

91 This change in the relation of weight to strength, according to the size of the structure, has long been familiar to engineers. (See, for example, “Comparisons of Similar Structures as to Elasticity, Strength, and Stability,” by Prof. James Thomson, Trans. Inst. Engineers, &c., Scotland, 1876.) The application to animal structures has been made by Herbert Spencer (Principles of Biology, Pt. II., ch. i.). The principle can be readily explained by models. Place a cubical block upon a square column. Double all the dimensions in a second model, which may be done by fitting together eight cubes like the first, and four columns, also the same as before except in length. Each column, though no stronger than before, has now to bear twice the weight.

92 Contractile force varies as sectional area of muscle. Let W be weight of Horse; w, weight of Bee; R, a linear dimension of Horse; r, a linear dimension of Bee. Then,

Contr. force of Hors eContr. force of Bee = sect. area of muscles (Horse )sect. area of muscles (Bee) = R²r².

But since Ww = R³r³, R²r² = Ww × rR.

Therefore Contr. force of Horse Contr. force of Bee = WrwR

But, by definition,

Rel. m.f. of HorseRel. m.f. of Bee = Contr. f. of HorseWContr. f. of Beew = Contr. f. of HorseContr. f. of Bee × wW =

WrwR × wW = rR = r³R^{3 13} = wW ¹³.

The weight of a horse is about 270,000 grammes, that of a bee ·09 gramme; so that

wW ¹³ = ·09270,000 ¹³ = ·000,000,3 ¹³ = ·0015 (nearly) =

Calculated Ratio of Relative Muscular Force of Horse to that of Bee. The Observed Ratio (Plateau) is ·523·5 = ·02128; so that the relative muscular force of the Horse is more than fourteen times as great in comparison with that of the Bee as it would be if the muscles of both animals were similar in kind, and the proportions of the two animals similar in all respects.

93 Rech. sur la Force Absolue des Muscles des InvertÉbrÉs. I^e Partie. Mollusques Lamellibranches. Bull. Acad. Roy. de Belgique, 3^e SÉr., Tom. VI. (1883).

Do., II^e Partie. CrustacÉs DÉcapodes. Ibid., Tom. VII. (1884).

94 Statical muscular force and Specific muscular force are synonymous terms in common use. Contractile force per unit of sectional area gives perhaps the clearest idea of what is meant.

95 Vol.II., p.203. The calculation here quoted is based upon an observation of Swammerdam, who relates that a Cheese-hopper, 1/4in. long, leaped out of a box 6in. deep.

96 Haughton’s Animal Mechanics, 2nd ed., p.43.

97 In any comparison it is necessary to cite not the height cleared by the man, but the displacement during the leap of his centre of gravity.

98 The granules are not shown in the figure, having been removed in the preparation of the tissue for microscopic examination.

99 Balfour, Embryology, Vol.II., p.603.

100 Yung (“Syst. nerveux des CrustacÉes DÉcapodes, Arch. de Zool. exp. et gÉn.,” Tom. VII., 1878) proposes to name connectives the longitudinal bundles of nerve-fibres which unite the ganglia, and to reserve the term commissures for the transverse communicating branches.

101 This commissure, which has been erroneously regarded as characteristic of Crustacea, was found by Lyonnet in the larva of Cossus, by Straus-DÜrckheim in Locusta and Buprestis, by Blanchard in Dytiscus and Otiorhynchus, by Leydig in Glomeris and Telephorus, by Dietl in Gryllotalpa, and by LiÉnard in a large number of other Insects and Myriapods, including Periplaneta. See LiÉnard, “Const. de l’anneau oesophagien,” Bull. Acad. Boy. de Belgique, 2^e SÉr., Tom. XLIX., 1880.

102 We have not been able to distinguish in the adult Cockroach the double layer of neurilemmar cells noticed by Leydig and Michels in various Coleoptera.

103 TraitÉ Anat., p.201, pl.ix., fig.1.

104 Phil. Trans., 1834, p.401, pl.xvi.

105 Vom Bau des Thierischen KÖrpers, pp.203, 262; Taf. z. vergl. Anat., pl.vi., fig.3.

106 The stomato-gastric nerves of the Cockroach have been carefully described by Koestler (Zeits. f. wiss. Zool., Bd. XXXIX., p.592).

107 “Mem. Acad. Petersb.,” 1835.

108 “Q.J. Micr. Sci.,” 1879, pp.340–350, pl.xv., xvi.

109 “Journ. Quekett Micr. Club,” 1879.

110 It is to be remarked that unusually large nerves supply the cerci of the Cockroach.

111 The number in Insects varies from eight to four, but seven is usual; four is the usual number in Crustacea.

112 “Q.J. Micr. Sci.,” 1885.

113 Exner has since determined by measurement and calculation the optical properties of the eye of Hydrophilus. He finds that the focus of a corneal lens is about 3mm. away, and altogether behind the eye.

114 Zur vergl. Phys. des Gesichtsinnes.

115 A critical history of the whole discussion is to be found in Grenacher’s “Sehorgan der Arthropoden” (1879), from which we take many historical and structural details.

116 Flies, whose eyes are in several respects exceptional, have almost completely separated rods, notwithstanding their quick sight.

117 Bull. de l’Acad. Roy. de Belgique, 1885.

118 References to the literature of the question are given by Hauser in Zeits. f. wiss. Zool., Bd. XXXIV., and by Plateau in Bull. Soc. Zool. de France, Tom. X.

119 Zeits. f. wiss. Zool., 1885.

120 Will confirms, by his own experiments (p.685), Plateau’s conclusion (Supra, p.46), that the maxillary and labial palps have nothing to do with the choice of food.

121 For a popular account of auditory organs in Insects, see Graber’s Insekten, Vol.I., page287; also J. MÜller, Vergl. Phys. d. Gesichssinn, p.439; Siebold, Arch. f. Naturg., 1844; Leydig, MÜller’s Arch. 1855 and 1860; Hensen, Zeits. f. wiss. Zool., 1866; Graber, Denkschr. der Akad. der wiss. Wien, 1875; and Schmidt, Arch. f. mikr. Anat., 1875.

122 Here, as generally in the digestive tube of the adult Cockroach, the peritoneal layer is inconspicuous or wanting. It occasionally becomes visible—e.g., in the outer wall of the Malpighian tubules, and in the tubular prolongation of the gizzard.

123 Plateau has expressed a strong opinion that neither in the stomach of Crustacea nor in the gizzard of Insects have the so-called teeth any masticatory character. He compares them to the psalterium of a Ruminant, and considers them strainers and not dividers of the food. His views, as stated by himself, will be found on p.131.

124 See Watney, Phil. Trans., 1877, Pt. II. The “epithelial buds” described and figured in this memoir are also closely paralleled in the chylific stomach of the Cockroach.

125 These epithelial buds have been described as glands, and we only saw their significance after comparing them with Dr. Watney’s account.

126 Development shows that these tubules belong to the proctodÆum, and not to the mesenteron.

127 The epithelial bands of the rectum of Insects were first discovered by Swammerdam in the Bee (Bibl. Nat., p.455, pl.xviii., fig.1). Dufour called them muscular bands (Rech. sur les OrthoptÈres, &c., p.369, fig.44).

128 “Lehrbuch der Histologie,” p.337.

129 Except in Dragon-flies and EphemerÆ.

130 Zeitsch. f. wiss. Zool., Bd. XXX.

131 The contents of the Malpighian tubules may be examined by crushing the part in a drop of dilute acetic acid, or in dilute sulphuric acid (10 per cent.). In the first case a cover-slip is placed on the fluid, and the crystals, which consist of oblique rhombohedrons, or derived forms, are usually at once apparent. If sulphuric acid is used, the fluid must be allowed to evaporate. In this case they are much more elongated, and usually clustered. The murexide reaction does not give satisfactory indications with the tubules of the Cockroach.

132 Bull. Acad. Roy. de Belgique, 1876.

133 Ib., 1877.

134 We are indebted to Prof. Plateau for the statement of his views given in the text.

135 Dissert. de Bombyce, pp.15, 16 (1669).

136 Biblia NaturÆ, p.410.

137 Schrift. d. Marburg. Naturf. Gesellschaft, 1823.

138 See, for a full account of this discussion, MacLeod sur la Structure des TrachÉes, et la Circulation PÉritrachÉenne (1880). The peritracheal circulation was refuted by Joly (Ann. Sci. Nat., 1849).

139 It may be observed that Graber, who has paid close attention to the heart of Insects, describes the inlets (e. g., in Dytiscus) as situated, not at the hinder end, but in the middle of each segment. We have not been able to discover such an arrangement in the heart of the Cockroach.

140 Lyonnet.

141 Brandt, Ueb. d. Herz der Insekten u. Muscheln. MÉl. Biol. Bull. Acad. St. Petersb. Tom. VI. (1866).

142 Arch. f. mikr. Anat., Bd. IX. (1872); Insekten, ch. x.

143 Newport, in Todd’s CyclopÆdia of Anatomy and Physiology, Art. Insecta, pp. 981–2.

144 Beitr. zur nÄheren Kenntniss von Periplaneta orientalis, p.19.

145 The termination of the aorta has been described by Newport, in Sphinx (Phil. Trans., 1832, Pt. I., p.385) Vanessa, Meloe, Blaps and Timarcha. (Todd’s Cycl., Art. “Insecta,” p.978.)

146 Moseley, Q.J. Micr. Sci. (1871).

147 The oldest Tracheate actually known to bear spiracles is the Silurian Scorpion of Gothland and Scotland (Scudder, in Zittel’s PalÆontologie, p.738). We need not say that this is very far removed from the primitive Tracheate which morphological theory requires. The existing Peripatus makes a nearer approach to the ideal ancestor of all Tracheates, if we suppose that all Tracheates had a common ancestor of any kind, which is not as yet beyond doubt.

148 The longitudinal air-tubes are characteristic of the more specialised Tracheata. In AraneidÆ, many JulidÆ, and Peripatus each spiracle has a separate tracheal system of its own.

149 Investigators are not yet agreed as to the minute structure of the tracheal thread. Chun (Abh. d. Senkenberg. Naturf. Gesells., Bd. X., 1876) considers it an independent chitinous formation, not a mere thickening of the intima. He describes the thread as solid. The intima itself is, he believes, divisible in the larger tubes into an inner and an outer layer, into both of which the thread is sunk. Macloskie (Amer. Nat., June, 1884) describes the spiral as a fine tubule, opening by a fissure along its length. He regards it as a hollow crenulation of the intima, and continuous therewith. Packard (Amer. Nat. Mag., May, 1886) endeavours to show that the thread is not spiral, but consists of parallel thickenings of the intima. He is unable to find proof of the tubular structure, or of the external fissure. We have specially examined the trachea of the Cockroach, and find that the thread can readily be unwound for several turns. It is truly spiral.

150 It has been supposed that these irregular cells of the tracheal endings pass into those of the fat-body, but the latter can always be distinguished by their larger and more spherical nuclei.

151 In the first abdominal spiracle the setÆ are developed only on that lip which carries the bow.

152 This subject is treated at greater length in Prof. Plateau’s contribution on Respiratory Movements of Insects. (Infra, p.159.)

153 Phil. Mag., 1833. Reprinted in “Researches,” p.44. Graham expressly applies the law of diffusion of gases to explain the respiration of Insects. Sir John Lubbock quotes and comments upon the passage in his paper on the Distribution of the TracheÆ in Insects. (Linn. Trans. Vol.XXIII.)

154 For an explanation of the physical principles involved in this discussion, and for the calculation (based upon our own assumptions), we are indebted to Mr. A.W. RÜcker, F.R.S.

155 J. Hutchinson, Art. Thorax, Todd’s Cycl. of Anat. and Phys.

156 De l’absence de mouvements respiratoires perceptibles chez les Arachnides (Archives de Biologie de Van Beneden et Van Bambeke, 1885.)

157 Ueb. d. Respiration der Tracheaten. Chemnitz (1872).

158 See table in Burmeister’s “Manual,” Eng. trans. p.398.

159 Art. “Insecta,” Cyc. Anat. and Phys., p.989.

160 Pogg. Ann. 1872, Hft. 3.

161 Works, Vol.IX., p.287. This passage has been cited by Rathke.

162 Arbeiten a. d. Zool. Zoot. Inst. WÜrzburg. Bd. II., 1874.

163 Phil. Trans., 1874, p.757.

164 The crystals have been supposed to consist of oxalate of lime (Duchamp, Rev. des sci. nat. Montpellier, Tom. VIII.). Hallez observes that they are prismatic, with rhombic base, the angles truncated. They are insoluble in water and weak nitric acid, but dissolve rapidly in strong sulphuric acid without liberation of gas, and still more rapidly in caustic potash. (Compt. Rend., Aug., 1885.)

165 It is usually stated that the spermatheca of the Cockroach opens into the uterus, as it does in most other Insects, but this is not true. Locusts and Grasshoppers have the outlet of the spermatheca placed as in the Cockroach; in other European Orthoptera, it lies upon the dorsal wall of the uterus. (Berlese, loc. cit., p. 273.)

166 It is a striking proof of the sagacity of Malpighi, that he should have observed in the Silkworm the spermatophore of the male (“in spiram circumvolutum persimile semen”) and the spermatheca of the female. His reasoning as to the function of the spermatheca wanted nothing but microscopic evidence of the actual transference of spermatozoa to establish it in all points. Audouin and Siebold supplied what was wanting nearly two centuries later, but they mistook the spirally wound spermatophore for a broken-off penis, and Stein (Weibl. Geschlechtsorgane der KÄfer, p.85) first arrived at the complete proof of Malpighi’s explanation.

167 The descriptions and figures of the reproductive appendages of female Orthoptera by Lacaze-Duthiers (Ann. Sci. Nat., 1852) are so often consulted, that it may be useful to explain how we understand and name the same parts. In pl.xi., fig.2, 8' and 9' are the 8th and 9th terga; the anterior gonapophyses are seen to be attached to them below; a (figs. 2 and 4) is the base of the same appendage, but the twisted ends are incorrect; the 8th sternum is seen at the back (figs. 2 and 4); a' represents the outer, f the inner pair of posterior gonapophyses.

168 We propose to notice here the chief differences which we have found between the figures of Brehm (loc. cit.), which are the fullest and best we have seen, and our own dissections.

Figs. 10, 11 (pp.169–70). The ejaculatory duct and duct of the conglobate gland are made to end in the penis (infra, p.178).

Figs. 14, 15 (p.173). These figures seem to us erroneous in many respects, such as the median position of the penis and titillator.

Fig.16 (p.174). The pair of hooks marked E are too small, and there are additional plates at the base, which are not figured (see our fig.102). F (of our fig.) is omitted.

169 In Blatta germanica the testes are functional throughout life. They consist of four lobes each. The vasa deferentia are much shorter than in P. orientalis.

170 The spermatocysts are peculiar to Insects and Amphibia. They arise by division of the spermatospores, or modified epithelial cells, and form hollow cysts, within which sperm cells (or spermatoblasts) are developed by further division. The sperm cells are usually placed radiately around the wall of the spermatocyst. They escape by dehiscence, and are transformed into spermatozoa.

171 Huxley, Anat. Invert. Animals, p.416.

172 The term “accessory gland,” used by Huxley and others, is already appropriated to glands which we believe to be represented by the utricles of the Cockroach, and which have only a general correspondence with the gland in question.

173 Similar organs, forming a male genital armature, have been described in various Insects. See Burmeister, Man. of Entomology, p.328 (Eng. Transl.); Siebold, Anat. of Invertebrates; Gosse in Linn. Trans., Ser. 2, Vol.II. (1883); Burgess on Milk-weed Butterfly, Ann. Mem. Bost. Soc. Nat. Hist.; &c.

174 In the following description it is to be understood that the observations have been made upon Blatta germanica, except where P. orientalis is expressly named.

175 Fertilisation consists essentially in the union of an egg-nucleus (female nucleus) with a sperm-nucleus (male nucleus). From this union the first segmentation-nucleus is derived.

176 Balfour, Embryology, Vol.I., p.337.

177 Q.J. Micr. Sci., Vol.XXIV., page596 (1884).

178 Kowalewsky in Hydrophilus, Graber in Musca and Lina, Patten in PhryganidÆ, myself in Meloe, &c.

179 Biolog. Centrablatt. Bd. VI., No. 2 (1886).

180 These terms are explained on p.115.

181 Cf. Korotneff, Embryol. der Gryllotalpa. Zeits. f. wiss. Zool. (1885).

182 In Gryllotalpa (Dohrn), as in Spiders, some Myriopods and Peripatus (Moseley, Phil. Trans., 1874), each stigma, with its branches, constitutes throughout life a separate system. The salivary glands arise in the same way, not, like the salivary glands of Vertebrates, as extensions of the alimentary canal, but as independent pits opening behind the mouth. Both the tracheal and the salivary passages are believed to be special modifications of cutaneous glands (Moseley).

183 Loc. cit.

184 This arrangement persists only in EphemeridÆ among Insects (Palmen, Ueb. paarigen AusfÜhrungsgÄnge der Geschlechtsorgane bei Insekten, 1884).

185 Genital pouch of the preceding description.

186 Indications, which we have not found time to work out, lead us to think that the development of the specially modified segments and appendages in the male and female Cockroach needs re-examination. We hope to treat this subject separately on a future occasion.—L.C.M. and A.D.

187 It may be useful to point out the following examples of parental care among animals in which, as a rule, the eggs are left to take care of themselves. It will be found that in general this instinct is associated with high zoological rank (best exemplified by Mammals and Birds), land or freshwater habitat, reduced number of eggs, and direct development.

188 The minute and early larvÆ of Toenia and Distomum may appear to contradict this statement. They really inhabit the film of water which spreads over wet grass, though they are capable of enduring dry conditions for a short time, like Rotifers and many Infusoria.

189 It is possible that the curious cases of agamogenetic reproduction of the larvÆ of Aphis, Cecidomyia, and Chironomus are vestiges of the original fertility of Insect larvÆ.

190 “Alia vero semen adhuc imperfectum et immaturatum recludunt, incrementum et perfectionem, sive maturitatem, soris acquisiturum; ut plurima genera piscium, ranÆ, item mollia, crustata, testacea, et cochleÆ: quorum ova primum exposita sunt, veluti origines duntaxat, inceptiones et vitelli; qui postea albumina sibi ipsis circum circa induunt; tandemque alimentum sibi attrahentes, concoquentes et apponentes, in perfectum semen atque ovum evadunt. Talia sunt insectorum semina (vermes ab Aristotele dicta) quÆ initio imperfecte edita sibi victum quÆrunt indeque nutriuntur et augentur, de eruca in aureliam; de ovo imperfecto in perfectum ovum et semen.”—De generatione, Exc. II., p.183 (1666). Viallanes justifies this view by applying it to the histolysis and regeneration of the tissues in Diptera. But these remarkable changes are surely secondary, adaptive, and peculiar, like the footless maggot itself, whose conversion into a swift-flying imago renders necessary so complete a reconstruction.

191 The reader is recommended to refer to Fritz MÜller’s Facts and Arguments for Darwin, especially chap. xi.; to Balfour’s Embryology, Vol.II., chap. xiii., sect. ii.; and to Lubbock’s Origin and Metamorphoses of Insects.

192 Those who care to see a bold experiment of this kind may refer to Haeckel’s SchÖpfungsgeschichte.

193 Comp. Embryology, Vol.I., p.451.

194 Yet none were so large as our largest living forms; their average size was very nearly that of Periplaneta americana.

195 Die Klassen und Ordnungen der Arthropoden. Leipzig, 8vo, p.292.

196 A few elytra of Coleoptera are recently announced from the Silesian “culm.”

197 Memoirs Bost. Soc. Nat. Hist., III., 23 seq. (1880).

198 See a paper on mesozoic Cockroaches now printing in the Memoirs Bost. Soc. Nat. Hist., Vol.III., p.439 seq.

199 The wingless creature from the Carboniferous deposits of SaarbrÜcken, described by Goldenberg as a Cockroach, under the name of Polyzosterites granosus, appears to be a Crustacean.

200 This includes all possible forms; our table shows but nine.