- 10 — 



The eggs of the gadoids fall into two groups, the one, which contains all the species 

 of the Gadus genus, having no oil-globule, whilst this is present in all the remaining 

 genera and species, where the eggs are known (Merlucciiis, Molva, Baniceps?, Onos, 

 Brosmius). The eggs of the gadoids are very "characterless", i. e. in the early stages at 

 least, where the embryonic pigment has not yet appeared, they lack those marks or char- 

 acteristics by means of which the separate species can be separated by simple observation, 

 as is the case with certain other species (e. g. Drepanopsetta, Clupea sprattus and pilchardus 

 etc.). The only expedient therefore, in determining preserved eggs in the early stages, is 

 to measure them under the miscroscope. It has proved namely by observation of ripe 

 eggs taken from spawning fish, that the eggs of the different species are not of the same 

 size, and the endeavour has therefore been made to determine the limits of variation of 

 the different species by measuring numerous eggs. There are however great difficulties to 

 contend against here; in the first place the limits of variation of the various 

 species overlap one another, and in the second place the eggs of the same species 

 may vary considerably in different regions. The result is that the difficulties 

 are often quite insurmountable in determining the eggs in the early stages; it is even 

 often impossible to separate a gadoid from for example a pleuronectid egg. From measure- 

 ments alone, in my opinion, there can be no question of absolutely certain results, unless 

 at least the collections are from restricted regions where the spawning 

 seasons and species are well known. 



The possibilities of successful determination are greater, if living eggs can be examined 

 and hatched out. In this way some species can be determined with certainty, for example, 

 Gadus caUarias and G.cpglefinus and Brosmius brosvie. Hatching has therefore been used 

 on a large scale oii board the "Thor", and I have not made a single determination 

 of a gadoid egg without hatching out the larva and could determine the 

 latter with certainty. 



Even the hatching method is however far from sufficient in all cases. For example, 

 if a gadoid egg gives rise to a larva which has yellow or both yellow and black pigment, 

 we cannot be certain in the present state of our knowledge whether the species is G. mer- 

 langus, minutus or Esmarki, or even not luscus or Poutassou. Nor can we decide 

 definitely whether a newly hatched larva is G. virens or pollachius, if we do not know 

 beforehand what waters the specimen came from. 



For these reasons I have only taken account of the eggs in a few cases, although 1 

 fully recognise the desirability of being also able to tabulate and map out on charts the 

 distribution of the pelagic eggs of the different species. I have believed it better in a work 

 like this only to give data which were absolutely certain and free from all 

 objections. This may be said of the determination of the postlarval stages of the 

 gadoids though not of the eggs, and I have therefore throughout only paid attention to 

 the former as the earliest stages in the life- histories of these fishes, which 

 can for the present be determined with certainty for all species. 



4. The use and importance of the material 



One of the chief objects with the present work was to determine the distribution of 

 the common European-Atlantic gadoids, several of which have great economical importance, 

 by means of uniform investigations over a large region. For this purpose 1 



