Eckinoderm Hybridization. 127 



(2) The average body length at 13 to 14 is greater than at 24 to 26. 



(3) The rods are longer than in the cold temperatures. (4) That the 

 number of cross connections in the anal rods increases. 



For Strongylocentrotus he notes with increased temperature: (i) Plu- 

 tei at 26 show an increase of multiple rods. (2) The average body length 

 at 24 to 25 is somewhat greater than at 13 to 14. (3) In infrequent 

 cases the beginning of lattice formation. 



SUMMARY OF THE RESEARCHES ON PURE FORMS. 



Summary of Vernon's observations. 



(1) Multiple anal arm rods sometimes appear in Strongylocentrotus. 



(2) The optimum temperature for fertilization, 17 to 22 C. 



(3) The optimum temperature for growth, 17 to 22 C. 



(4) The optimum temperature for arm growth, 17 to 22 C. 



(5) The optimum light for growth is ordinary diffuse daylight. 



(6) There is a species temperature optimum for Strongylocentrotus, 



Sphatrechinus, and Echinus. 



Summary of Herbst's observations. 



(1) Increased temperature gives an increase in multiple rods. 



(2) There is a species temperature optimum for body size. 



(3) In Strongylocentrotus and Echinus, in a small percentage of 



cases, lattice formation is caused by increased temperature. 



The outcome of these researches has been to show that some of the 

 varieties of skeletal structure found in hybrids may occur as normal 

 variations in purely bred larvae, and further that such variation may be 

 induced by a change in the external conditions, notably by changes in 

 temperature. 



After considering the cross-fertilization results I shall refer again 

 to these observations on purely bred forms, when their importance will 

 be evident. 



OBSERVATION ON CROSSES. 



Vernon (1898, 1900) working with Arbacia, Dorocidaris, Echino- 

 cardium cordatum, Echinocardium mediterraneum, Echinus acutus, 

 Echinus microtuberculatus, Sph&r echinus, and Strongylocentrotus, out of 

 64 possible direct and cross-fertilizations tried 49. Of these, 29 gave 

 plutei of 8 days' growth, 9 gave segmenting ova, blastulae, or gastrulae, 

 and 1 1 gave no sign of cross-fertilization whatever. 



It is in a measure unfortunate that the fine observations of Vernon's 

 earlier paper should be based on body proportions alone. This situation 

 is to some extent relieved by the supplementary and the new observa- 

 tions on the skeletal structures described in the later paper. 



Perhaps the most important of Vernon's results was the determi- 

 nation of a seasonal variation in the character of the plutei obtained 

 from the crosses. In general, with Echinus- Arbacia, Echinus-Strongy- 

 locentrotus, Sph&rechinus-Echinus, and Sphcerechinus-Strongylocentrotus, 

 i.e., with the most successful crosses, the summer larvae were of the 

 maternal type, while the autumn and winter larvae were of the paternal 



