* The origin of these branches—the dermato-branchial connectives—is marked in some specimens by white spots upon the primary horizontal bars.

* See p. 132.

* See p. 132.

* The walls of the slit were definite, straight, and smooth, resembling in all respects those of the slit in Ascidia mentula. It must not be imagined that the slit, which I have mentioned, was an irregular abnormality of the kind described by Prof. Herdman in specimens of Ascidiella aspersa from the west coast of Ireland (Proc. Liv. Biol. Soc., v, 1891, p. 210, pl. x), an abnormality which may also occur in Ascidia mentula, as I have myself observed in a specimen from Loch Long.

* This distinction of size is much less apparent in mature than in young individuals.

* It is needless to say that we look forward with interest towards Prof. Herdman's promised re-description of some of Alder and Hancock's types.

* Journ. Linn. Soc. Zool., vol. xxiii, 1891, p. 594.

* I give this name to the curious aperture, so commonly found in the pharyngeal wall of Ascidia mentula, in which species it was first noticed by Kupffer (1. c.). It has been ingeniously suggested lately that it represents the persistent internal opening of the right primitive atrial canal, in spite of the fact that it is absent in the more primitive Ascidians, such as Clavelina and the Distomidœ. Now, as has been stated above (pp. 123 and 124), I have discovered this slit to be present in large individuals of two other species of Ascidians which are not closely allied to Ascidia mentula (Ascidiella aspersa and Ascidia mollis), although it does not exist in young specimens of those species. This fact is a sufficient disproof of the theory which gives to the slit the value of a phylogenetic remnant. My own theory is less attractive, but possibly more true. The slit is always situated opposite the cloacal orifice, and only occurs in large species (Ascidia mentula and its close allies, e.g. Ascidia lata, Herdman) and in large individuals of smaller species (e.g. of A. mollis and Ascidiella aspersa). May it not be a special adaptation for the prevention of the over-accumulation of fæces in the cloacas of large Ascidians, where the ordinary methods of ejection are insufficient? Ascidians, being sessile animals, are especially liable to danger from such over-accumulation, as Giard long ago stated in the case of the Didemnidæ and Polyclinidæ (Arch. Zool. Exp., i, p. 520); and special means are adopted in various sections of the group to ward off the danger. For instance, as Maurice has well suggested, the cloacal languettes of the Polyclinidæ serve the definite function of keeping open the cloacal canals in colonies of that family (Arch. de Biol., viii, 1888, p. 243); while in the Botryllidæ the end is achieved only by the united efforts of the zooids in a cœnobium: they simultaneously and suddenly contract their bodies, and so drive a strong current of water through their peribranchial cavities into the common cloaca, ejecting the fæces with such violence, as Gaertner observed, “ut ingenti saltu oppositum faveæ marginem transiliant” (see Giard, loc. cit.).

In the large Ascidians under discussion, the presence of this big oval slit—it is frequently over a centimetre in length—directly opposite the cloacal cavity, will enable the animal, by a strong contraction of the muscular tunic, to drive a considerable body of water from the pharynx into the cloaca, and thus to effect the desired object more thoroughly than is possible when stigmata exist alone.

Kupffer has also recorded the existence of paired pharyngo-atrial slits, symmetrically placed in the posterior region of the pharynx, in Ascidia conchilega and Ciona [canina] intestinalis. The former species I have been unable to examine, but in C. intestinalis (preserved material) some individuals posses huge slits, through which the intestine conspicuously projects into the pharynx, while in other individuals no unusual apertures can be made out at all. (Cf. Traustedt, loc. cit., p. 455. Heller, loc. cit., ii, p. 118, seems merely to repeat Kupffer's statement. Roule, loc. cit., makes no reference to any exceptional openings.) I am inclined, therefore, to believe that in both these species Kupffer's apcrtures are accidental or artificial rather than natural.

* Since the above was put in type, I have been enabled to examine some specimens of A. mentula, which were dredged in Loch Long and are now under Mr. Hoyle's charge in the Manchester Museum. The number of tentacles is so variable as to be only 18 in an individual 4¾ inches long, while it is nearly 40 in an individual 3 inches long.