chap, vi.] ANNELIDS : RECAPITULATION. 167 



ticular sacs which communicate with a tortuous vas deferens 

 on each side which together enter a single penis. The paired 

 ovaries are placed behind this and the oviducts unite to form 

 a common vagina. 

 121. In a specimen found amongst a large series investigated, each 

 vas deferens opened by a separate penis, of which the most an- 

 terior opened in the 20th annulus and the posterior in the 25th. 

 The female apparatus was similarly divided. One ovary was 

 placed near the penis in the 25th annulus and from it a vagina 

 passed down to open with the penis. The other ovary, with a 

 similar vagina, lay in the 30th annulus. Asper, G., Zool. Anz., 

 1878, I, p. 297. 



Recapitulation of evidence as to Oligoch^eta and Hirudinea. 



Variation in these two groups appears in such similar modes 

 that points of special consequence in both may conveniently be 

 spoken of together. 



1. As elsewhere seen, so here, there are forms, e.g., Perionyx 

 excavatus or Pachydrilus sphagnetorum, shewing great variability, 

 while others, the common Earthworm for instance, rarely vary. 



2. Both forward and backward Homceosis may occur ; a form 

 normally having the </ pores, for instance, on the 15th segment, 

 may as an individual variation have them on the 16th (No. 105), 

 while an individual of another genus, starting from the same 

 normal, may have them on the 13th (No. 111). 



3. As in other cases of Homceosis, when a member of a 

 Meristic Series, in this case a segment, develops an organ proper 

 to another segment, this organ is formed in a place serially 

 homologous with its normal place. (To this principle certain 

 limitations must hereafter be introduced.) 



4. Variation may, or may not, be simultaneous and cor- 

 related in the several systems. The position of the J openings 

 for example, may or may not vary similarly and simultaneously 

 with that of the £ openings, though on the whole the evidence 

 suggests that such correlation is not uncommon. The facts seen 

 in the genus Allurus, in which one species (A. tetrdedrus) has the 

 c/ 1 pore normally in front of the $ pore, sufficiently indicate that 

 the variation in the position of these two openings is not always 

 so correlated. It may be further mentioned that variation in 

 number of ovaries seems to occur generally without correlated 

 variation in the number of oviducts. 



5. Such Variation may or may not be simultaneous on the 

 two sides of the body. When not thus bilaterally symmetrical, 

 there may nevertheless be a full correlation between the parts 

 of the same side. 



6. The evidence does not indicate any limit to the number 

 of segments which may take on a certain character, or approxi- 



