188 ME. F. J. COLE OX TJIE STRUCTUEE AND MOEPHOLOGY OF 



Emery, the pit organs of Allis, and the " Spalt-papillen " of Fritsch and Ewart & 

 Mitchell are the same in every structural detail, it cannot be denied (1) that all these 

 organs belong to the lateral line system, and (2) that they are all varieties of one form 

 of sense organ representing the first stage in the phylogeny of a sensory canal. 



Merkel's work (1880, 138), published independently of Emery's, places the whole 

 question on a solid basis of fact. He showed that there were several kinds of epidermal 

 pits, all differing but slightly from each other, and yet roughly characteristic of the 

 different groujisof fishes. His sharp distinction between the " Endknospen " or terminal 

 buds, which do not belong to the lateral line system at all, and the sensory pits, which 

 do, has been of the utmost service in determining the homology of the latter in the 

 various groups of fishes. He further pointed out that the presence of these pit organs 

 is correlated with the occurrence of other sense organs of the lateral line system, 

 i. e. that they were most numerous in tliose Teleostean fishes Avhere the canal organs 

 were reduced in number, and practically absent in the Elasmobranch fishes, cohere they 

 are replaced by the ampvlhe of Lorenzini. The importance of tliis fact will be 

 emphasized further on. Sappey (1880, 175) makes some curious mistakes. He con- 

 siders Lorenzini's ampullae to be " glandes," bvit distinguishes between the ampullary 

 system and the canal system — the latter having no " glandes" and consisting of canals 

 opening on to the surface by a variable number of tubules ( = dermal tubules). Sappey 

 believed the "non-glandular" sensory canals to be probably tactile in function. 

 It is interesting to note that Solger considered a line of pit organs to represent a 

 potential canal, and he describes (1880, 193) the lines of pit organs accompanying the 

 lateral or body canal as accessory lateral lines. 



Balfour's views on the question are in many respects prophetic, and of such importance 

 as to be quoted hi extenso. He says (1881, 9, vol. ii. p. 445) : " It is clear that the 

 canal of the lateral line is secondary, as compared with the open groove of Chimaira or 

 the segmentally arranged sense bulbs of young Teleostei ; and it is also clear that the 

 l)hylogenetic mode of formation of the canal consisted in the closure of a primitively 

 open groove. The abbreviation of this process in Elasmobranchii was probably acquired 

 after the appearance of the food-yolk in the q^q,, and the consequent disappearance of 

 a free larval stage. While the above points are fairly obvious, it does not seem easy to 

 (lecde a priori whether a continuous sense groove or isolated sense bulbs were the 

 ])rimitive structures from which the canals of the lateral line took their origin. It is 

 equally easy to picture the evolution of the canal of the lateral line either from (1) a 

 continuous unsegmented sense line, certain points of which became segmentally dilferen- 

 tiated into special sense bulbs, while the whole subsequently formed a groove and then 

 a canal ; or from (2) a series of isolated sense bulbs, for each of which a protective 

 groove was developed, and from the linear fusion of wliich a continuous canal became 

 formed*. Erom the presence, however, of a linear streak of modified epidermis in 

 larval Teleostei, as well as in Elasmobranchii, it appears to me more probable that a 



* Siiiee this was written, Allis has described the lateral canals of Anim as arising in jireeisely the manner here 

 described (i. e. 2). 



