240 BULLETIN OF THE 



or closely related to Agalma, to know accurately the changes in external 

 form which the ovum passes through in these genera. Upon such knowl- 

 edge we can hang our speculations regarding the possible descent of the 

 members of the Siphonophora one from another, or from a common 

 ancestor. 



The species of Agalma which has been studied is the only Agalma 

 thus far recorded from New England waters. It is called Agalma ele- 

 gans, and was first described by the author. When this animal was first 

 taken, in 1876, I regarded it as the same as the "form (b) " of Agal- 

 mopsis elegans Sars, or closely related to it. 



Early Changes of the Egg before Segmentation. 



The earliest changes in the egg take place, in all cases observed, 

 while it is enclosed in the female gonophore.* These go on with great 

 rapidity, as will be shown by the following statements. Specimens of 

 Agalma captured on August 6, at noon, were found four hours later to 

 have dropped their gonophores, from which had come ova segmented 

 in the 4-cell stage. It must be mentioned, however, that by transfer- 

 ring the Agalma from the sea into aquaria they were placed in unnatural 

 conditions, so that changes in temperatiu'e and other causes may have 

 accelerated or retarded their rate of growth. There is nothing to show 

 that there are not other kinds of segmentation besides that which is 

 he^e described. 



* In a popular article on the development of Agalma clcgans, published in the 

 American Naturalist for March, 18S1, certain changes in the gerniinative vesicle 

 which were mistaken for segmentation were spoken of. This interpretation was 

 erroneous, and the true segmentation was not described. On p. 188, op. cit., the egg 

 is spoken of as cast into the water and there impregnated. There is nothing to prove 

 that this is the case in Agalma. It has, however, excellent support in the history 

 of observation. Gegenbaur, Beitrage zur naheren Kenntniss der Schwimmpolj-pen 

 (Siphonophoren), p. 49. Writing of the genera Agahnopsis, Forskalia, Phijsophora, 

 Hippopodius, and Diphycs, he says : "Die Befruchtung erfolgt erst nach deni Aus- 

 tritte der Eier aus der Eikapsel ; denn niemals fand ich Samenfiiden in letzere einge- 

 drungen, ebeu ausgetretene Eier dagegen stets von ihnen unischwarmt. Sie sassen 

 dann strahlenartig mit dem Kbpfchen an der Peripherie des Eies an, mit dem Faden- 

 theile selbst in zitternder Bewegung. " I have not been able to observe a similar 

 condition in Agalma, nor was a free egg with nucleus and nucleolus found floating 

 in the water. In one instance these bodies were observed to vanish while yet the 

 ovum was in its gonophore, while the stalk of the same was attached to the parent. 

 All eggs found free from the gonophore are destitute of these structures. 



