— i8o 



The first time these eggs were found was on May 22nd 1904 in the Atlantic to the 

 west of the Faeroes (6l"l5' N., 9°35' W.), where the depth varied from 970 to 872 

 meters. The following hauls with the young-fish trawl were made here: 



1. Young-fish trawl, 15 meters wire out, 45 minutes: no fish eggs. 



2. do. 100Ü meters wire out (not on bottom), 1 hour: ca. 20 large fish eggs 



most of which with large embryos. 



3. do. 1700 meters wire out (on bottom), 1 hour: ca. 30 large fish eggs as in 



haul No. 2, but fewer with so large embryos. 



These large eggs' from this station, which resemble a good deal, the muraenoid 

 eggs described by Raffaele and by Eigenmann, belong to the deepwater salmonoid Argen- 

 lina silus, the larval and postlarval stages of which are represented in great quantities 

 in the "Thor's" collections (from the same hauls which gave the large bathypelagic eggs). 



They are mainly characterised by the folio ving features: (l) their large size (about 

 3 mm. in diameter); (2) the presence of a very large (ca. l mm.) oil- globule, which 

 was yellow, colourless or pinkish. In addition to the large globule there were some- 

 times also a few smaller ones. (3) by the yolk, which is very large, especially in 

 the younger eggs in which it fills the whole egg (see egg No. 2 in note below), and 

 which is not homogenous but is of a fine vesicular or granular structure as has 

 been described in muraenoid eggs ; (4) the embryo present in some of the eggs had the 

 anus close to the end of the tail (cf. Clupeidae) and opening at the edge of the 

 broad embryonic fin. The larvae just hatched were quite lacking in pigment. The 

 number of preanal myomeres was 46—48 or exactly the same as I have found in larval 

 and postlarval specimens of Argentina silus. 



These large eggs were pelagic, as was shown amongst other things by the fact 

 that they came up in haul No. 2, when the net had not been at the bottom (as is well- 

 known, the presence or absence of bottom material easily determines whether the net has 

 been at the bottom or not). Further, they were bathypelagic, as we only got them 

 in the deep hauls not on the surface, and this was also the case on many later occasions. 



Along with the hydrographer on the "Thor", Cand. mag. J. N. Nielsen, I made the 

 following experiments. .Some of the large fish eggs which were still living were placed 

 into surface water with a specific gravity of 1-02738. In this the most of them sank, only 

 a few remaining floating. The eggs which had sunk in the surface water were then 

 placed in water with a specific gravity of l -02870 (surface-water with an addition of salt), 

 in which they all floated. Some of them floated already in water of 1.02831 sp. gr. 



1 In 7 of these eggs the following measurements (in mm.) were found: 



Diameter of egg 



— - yolk 



Largest diameter of oil-globule 



No I 



3-24 

 2.50 

 1.16 



No: 



No 3 



No 4 



No 5 



No 6 



34s 

 3-46 

 1.09 



3-2S 

 2.1 1 

 1. 16 



3-34 

 2. 1 1 

 1.09 



3-41 

 3-14 

 1.14 



3.20 

 2.52 

 1.09 



No 7 



3-07 

 2.32 



0-9S 



In 8 embryos taken from the eggs the following number of preanal myomeres were found, 46, 46, 46, 46, 

 47, 47, 47, 48 (the last preanal myomere was considered to be the one lying over the place where the end of 

 the gut bends out from the body to open at the edge of the embryonic fin). 



