The manner in which parental characters are transmitted to the offspring when different species of Echinoderms are crossed has been the subject of much experimental enquiry and quite contradictory conclusions have been arrived at by different investigators. Thus Vernon (13), who carried out a most extensive series of experiments with the species of the genera Arbacia, Echinus, Strongylocentrotus, Sphærechinus, and Echinocardium which are available at Naples, came to the conclusion that the condition of the genital glands of the parents (whether imperfectly ripe, fully ripe, or stale) determines in many cases whether or not a hybrid will be formed, and further that though in the majority of cases the hybrid exhibits purely maternal characters, yet it sometimes exhibits paternal characters also, and that this result is also due to the condition of ripeness of the genital glands of its parents. Herbst (5), who also worked at Naples and who used the genera Echinus, Strongylocentrotus, and Sphærechinus for his experiments, found also that the hybrids in many cases showed the paternal influence, but that the extent to which this influence was exhibited varied with the temperature. Doncaster (1), who likewise worked at Naples, also arrived at the conclusion that the greater or less development of paternal characters in the hybrid was due to the temperature. On the other hand, Loeb (7, 8) and his pupil Hagedoorn (4) came to the conclusion that the hybrid exhibited purely maternal characters, and Fischel (2) arrived at the same conclusion on the whole. This conclusion is the more remarkable because Hagedoorn in his experiments used two species of the same genus. Tennent (12) crossed species of the American genera Toxopneustes and Hipponoe and found that the characters of Hipponoe were dominant in the hybrid whichever way the cross was made, but that if the alkalinity of the sea-water were reduced by the addition of dilute acid the influence of Toxopneustes became increased. Lastly Loeb, Redman King, and Moore in a joint paper published quite recently (9), in which they record the results of experiments with the same two species which Hagedoorn used, arrive at the conclusion that the exhibition of paternal and maternal characters in the hybrid is governed by the principle of Mendelian dominance, since, as they assert, the same characters appear in the hybrid whichever way the cross is made, whether, that is to say, in any particular case the character in question is inherited from the male or from the female parent. During a study of the whole subject which I recently made with the object of summarising the present state of our knowledge of this question of the inheritance of paternal and maternal characters in the hybrid, I was struck with the necessity of a preliminary thorough investigation of the characters of the normal larvæ of the species used in hybridisation experiments. The amount of general acquaintance with Echinoderm larvæ displayed by several of the investigators who have attacked the subject is, to say the least, somewhat defective. Thus Herbst (5), who studied chiefly the cross between Sphærechinus and Strongylocentrotus, attaches great importance to the extent to which lattice-work appears in the skeleton of the arms of the hybrid. In the normal larva of Strongylocentrotus, it is true, all four arms are supported by unbranched calcareous rods, whilst in the normal larva of Sphærechinus, each of the two posterior arms is supported by parallel rods connected by cross-pieces like the steps of a ladder, an arrangement which is termed “lattice-work.” But Herbst fails to take into account the fact that in the normal larva of Strongylocentrotus a lattice-work skeleton can appear as a variation, and hence an attempt such as he makes to estimate quantitatively the influence of one parent by the amount of lattice-work which appears in the hybrid rests upon an insecure foundation.
InfPolyn, a Nonparametric Bayesian Characterization for Composition-Dependent Interdiffusion Coefficients