260 TEST ACE A ATLANTIC A. 



In a Bubfossil condition, the L. tornatellina is tolerably com- 

 mon at Canipal; but I am not aware that it has been ob- 

 served in the deposits of either Porto Santo or the Bugio. 



Although regarded hitherto as quite peculiar to the Madei- 

 ran archipelago, I may just add that a single example of it, which 

 I have inspected with the greatest care, was met with by Mr. 

 Watson, a few years ago, in Grand Canary. 



Lovea mitriformis. 



Achatina mitriformis, Lowe, Ann. Nat. Hist. ix. 120 (1852) 

 Azeca ? mitriformis, Pfeiff., Mon. Hd. iii. 522 (1853) 

 Achatina mitriformis, Lo^ve, Proc. Zool. Soc. Lond. 203 



(1854) 

 Glandina Alb., Mai. Mad. 59. t. 15. f. 15, 16 



(1854) 



Achatina tornatellina (pars), Paiva, Mon. Moll. Mad. 112 



(1867) 



Habitat Maderam Desertasque tres (prsesertim has), vulga- 

 ris : in Portu Sancto rarissima, sc. semel tantum, in summo 

 monte ' Pico Branco ' dicto, adhuc lecta. In statu semifossiii 

 in ins. Deserta Australi invenitur. 



It is far from impossible that this may be in reality but a 

 rather narrow and slightly elongated form of the L. tornatellina, 

 in which the subvertical ventral plait is more appreciably denti- 

 form ; and all the more so, since the examples of it in Madeira 

 proper have the latter less abruptly expressed than in those from 

 the Desertas and Porto Santo. Nevertheless since the shell has 

 undoubtedly a different outline and facies, and it was the opinion 

 of Mr. Lowe that it is not absolutely con specific with the torna- 

 tellina, I will not attempt (as the Baron Paiva has done) to 

 suppress it. Indeed the mere fact of its occurring on the whole 

 five islands of the Grroup, and often in company with the torna- 

 tellina, would supply a certain amount of at any rate presump- 

 tive evidence that it is at least no local phasis of that universal 

 species. But these abstract questions of ' species ' and ' variety ' 

 are so difficult (indeed in many instances so impossible) to solve, 

 that where any given form is sufficiently well defined to be 

 easily recognisable, and it is not connected with its nearest ally 

 by decided intermediate links, I prefer for my own part (at 

 any rate in those instances where it has already been enunciated 

 and published) to accept it as specifically distinct. 



With these remarks I may add that the L. mitriformis 

 is separable from the tornatellina, mainly, by its relatively 

 somewhat narrower and more elongate outline (the form being 

 less ventricose, the spire a trifle more produced, and the suture 



