54 ME. T. J. PAEKEE OX THE INTESTINAL 



valve as that shown in fig. 3, PI. XI., in which there is again a " nest" of cones, but 

 with their apices directed forwards instead of backwards, and with their altitudes 

 approximately equal. All the turns have, indeed, the form and direction which is 

 normal to the first, and which it retains even in extreme cases of backward defiection. 



§ 7. There are thus four chief forms assumed by the spiral valve of Baia, the cha- 

 racters of which may be recapitulated as follows : — 



Type A. Simplest form : free edge in all the turns, but the first on the same level 

 as the attached edge ; widtla of valve not greater than half the semidiameter 

 of the intestine. This form is more or less hypothetical, the nearest undoubted 

 approach to it being that shown in fig. 1, PI. X. 

 Ty2)e B, Regular spiral : tlie free edge is again on the same level as the corre- 

 sponding portion of the attached edge ; but the width of the valve is equal to 

 the semidiameter of the intestine, so that its thickened free edge forms a 

 columella (fig. 4, PI. X.). 

 Type C. All the turns but the first are deflected backwards, the width of the valve 

 becoming much greater than the semidiameter of the intestine (fig. 8, PI. X., 

 and fig. 1, PI. XI.). 

 Type B. All the turns deflected forwards, the width of the valve again being con- 

 siderably greater than the semidiameter of the intestine (fig. 3, PL XI.). 

 It will be seen that, of these four types, A forms a starting-point or stem form, from 

 which the other three may be supposed to have diverged by the turns of the valve 

 growing either directly inwards {B)^ or backwards (C), or forwards {!)) during their in- 

 crease in width. And this view is borne out by the occurrence of intermediate stages. 

 I have found one or two undoubted transition-forms between A and B ; and by far the 

 greater number of the valves examined are intermediate either between A and C or 

 between A and I). 



§ 8. I have figured two of these intermediate forms, figs. 6 & 7, PI. X., which show 

 in a very striking manner that the variations in the valve are quite independent of 

 species or sex. Both these specimens, in fact, were taken from adult male examples of 

 B. maculata of as nearly as possible the same size. 



To show that the ■\'ariations are wholly independent of the age of the animal, I may 

 mention that although the simplest form (fig. 1, PI. X.) is from a small specimen, yet 

 one of the most complicated (fig. 8, PI. X.) is from one very little larger, while a far 

 simpler form is often seen in full-grown animals. Moreover I have examined two or 

 three specimens of Ray not more than three inches long, excluding the tail, and found 

 that in them the valve exhibited all the variations occurring in the adult, one approach- 

 ing very closely to type C, another to B. 



^ 9. It will at once be perceived that the variations just described in the form of the 

 spiral valve will produce a corresponding diff'erence in the amount of absorption-surface 



