ENDOSKELETAL SYSTEMS OF LIMULUS AND SCORPIO. 367 



appear to me to correspond to the internal branchials (48) of the four corresponding- 

 pairs of limbs in Limulus. 



The view which I advocated in my essay " Limulus an Arachnid," as to the mode of 

 conversion of an external lamelligerous appendage into the hollow lamelligerous lung 

 of Scorpio, no longer commends itself to me. A much simpler and, as it appears to me, 

 a thoroughly satisfactory explanation of the relationship of the two organs has suggested 

 itself in the course of the investigations here recorded, and is supported also by 

 embryological data. In the essay above referred to, I suggested that by the enlarge- 

 ment of the hollow stigmata connected with the thoraco-branchial muscles of an 

 ancestral Scorpion, resembling Limulus in having branchigerous appendages on the 

 mesosoma and thoraco-branchial muscles, the branchigerous appendage might come to 

 lie in the pit or hollow of the tendon, and eventually the hollow might enclose it. 

 The conversion of the in-sunken appendage into a hollow air-holding sac and the 

 corresponding conversion of the surrounding pit into a closed blood-holding space, 

 involved serious difficulties which were indeed fatal to the hypothesis. When I found 

 that the muscle (veno-pericardiac) attached to the apex of each lung-sinus in Scorpio 

 had no possible relation to the thoraco-branchial muscles of Limulus, but was repre- 

 sented in Limulus by exactly similar veno-pericardiac muscles, I gave up my over- 

 strained hypothesis. 1 trust that the failure of my previous suggestion will not unduly 

 prejudice those interested in this subject against that which I now advance. Since 

 my memoir "Limulus an Arachnid," Dr. MacLeod of Brussels has published some 

 speculations on this subject, in which he puts forward an ingenious theory of his own 

 as to the mode in which the lamelligerous appendage of a Limulus-\\ke animal might 

 be converted into the lamelligerous lung-book of an Arachnid. I will not enter into a 

 discussion of Dr. MacLeod's hypothesis, but will merely point out that inasmuch as it 

 deals with not the less modified lung-book of Scorpion, but the more modified lun<*- 

 book of the Araneina, it is unsatisfactorily elaborated. The lung- book of Sco?pio has 

 a definite axis carrying the leaf-like lamella?, and corresponding to the axis of the same 

 animal's pecten. Such an axis is not present in the Araneine lung-book, and yet must 

 be accounted for as a primary structure in any theory as to the origin of these organs. 



The hypothesis which I now put forward is perfectly simple, and leaves, I think, 

 nothing to be desired. In Limulus, as in Scorpio, there is on each side of the sternal 

 surface a great blood-sinus in free communication with the lamelligerous organs. Let 

 us suppose such to have been the case in the common ancestor of these two animals, 

 and let us suppose that this ancestor possessed six pairs of mesosomatic appendages, of 

 which five were lamelligerous and intermediate in form between the pectens of Scorpion 

 and the recent Limulus appendage. Now suppose that in the Scorpion branch of the 

 family the mesosomatic appendages grew relatively smaller and smaller, were no longer 

 locomotor organs, but purely respiratory, and served for aerial rather than aquatic 

 respiration. If we imagine the four hinder pairs of these reduced appendages to have 



vol. xi. — paet x. No. 8. — May, 1885. 3 r 



