416 DR. E. A. GOELDI ON THE LEPIDOSIBEN OF THE AMAZONS. 



brown. There is, however, little doubt that this is largely, if not altogether, due to the 

 longer action of alcohol and sunlight upon the Brazilian specimens." But this should 

 be taken in connection with my observations on the subject taken in loco and from 

 fresh material. 



V. Segmkntation of Limb-axis. 



I have already had occasion to state my opinion upon this question in my letter to 

 Dr. A. Giinther, published in 'Nature' (liv. p. 270). In the first individual sent 

 to me from Marajo, I discovered that the segmentation, considered by Prof Ehlers to 

 constitute a peculiarity of his Paraguayan Lepidosiren articulata, exists also in the 

 Amazonian Lepidosiren, as clearly shown by Prof. Lankester in the above-mentioned 

 memoir. The segmentation of the cartilaginous limb-axis evidently escaped the notice 

 of the first describer Bischoff", but this oversight does not materially detract from the 

 value of the excellent memoir upon Lepidosiren issued in 1840 by the celebrated 

 anatomist of Vienna. 



VI. System op Lateral Lines. 



It seems that nobody has added anything more to the knowledge of the lateral-line 

 system since Bischoff's memoir, and that anatomist only mentioned and figured the 

 lateral line of the head-region of Lepidosiren. The recent figures of Lepidosiren and 

 Protopterus contained in Prof Lankester's memoir show what I consider to be a 

 median lateral line on each side of the body ; but in the text of his article no allusion 

 or description is found referring to this point. Although surprising, however, this 

 circumstance may be easily understood. Dead specimens of Lepidosiren generally 

 become rapidly dark and the lateral lines cannot be so distinctly perceived, as I have 

 had occasion to verify in my own spirit-specimens. Fig. 2, PI. XXXVII. , taken from 

 our living Amazonian Lepidosiren, shows most distinctly six lateral lines — that is, 

 three on each side of the body, running longitudinally and in a straight line from 

 head to tail. 



a. Median line. 



I. Dorsal line. 



c. Ventral line. 



The best-developed line, being nearly uninterrupted, is the median, giving out from 

 time to time small vertical branches, generally two together, towards the dorsal line. 

 The dorsal line is less developed, being composed of longer or shorter, straight or 

 curved marks, the longitudinal and lineiform arrangement of which is recognized 

 witlioiit difficulty despite the numerous interruptions. It seems that these marks are 

 small lateral branches, similar to those of the median line, but directed downward in 

 the places where the dorsal line remains better preserved. Very well developed is also 

 the ventral line, emitting vertical branches downward, nearly reaching the true ventral 

 margin and almost touching the corresponding branches of the other side. When our 



