338 BULLETIN OF THE BUREAXX OF FISHERIES 



inshore migration) takes place from February until near the end of May; and 

 the evidence afforded by our tow nettings, if not conclusive, suggests an equally 

 protracted spawning season in the Gulf of Maine, for on the one hand we have 

 taken larviE already 27 mm. long as early as May 10, and on the other, newly 

 hatched larvse (only 6 to 7 mm.) as late as June 19 in the inner parts of the Gulf 

 and as late as July 23 on Georges Bank. In fact we took one only 10.5 mm. long 

 on August 22 in 1912 off Seguin Island. In the Gulf of St. Lawrence,. however, 

 where vernal warming is later than in the Gulf of Maine, lumps probably do not 

 commence spa^vning until the middle of April, for Cox and Anderson found no 

 larvae imtil late in June, their observations pointing to' late May as the height of 

 the breeding season there and to mid-Jime as about its termination, °' assimiing 

 that the period of incubation is about as long in the Gulf of St. Lawrence as in 

 European waters of like temperature — that is, six weeks to two months. 



On the other side of the Atlantic spawning takes place in very shallow water 

 chiefly close to low-tide mark, but the fact that the egg masses (more or less familiar 

 objects on European shores) seem never to have been reported along the coast of the 

 Gulf of Maine, although the local presence of larvae is proof that lumps breed all 

 around its periphery, suggests that the eggs are deposited at least a fathom or two 

 down. Our capture of recently hatched larvae over Georges Bank is evidence that 

 the latter also serves as a spawning ground in 15 to 25 fathoms or deeper, but the 

 lower limit to spawning is yet to be determined. 



Females are prolific, large ones of 18 inches producing up to 136,000 eggs 

 which sink and stick together in large spongy masses through which the water 

 circulates freely. In north European waters these egg masses are often found 

 adhering to rocks or other objects or in crannies near low-water mark, and watch 

 should be kept for them along the rocky coast line of the Gulf of Maine. The 

 male lump, like the sticklebacks, guards the eggs until they hatch, his courage and 

 his devotion to his charge having often been described."* Throughout the period 

 of guardianship, which he performs fasting, he constantly fans the egg mass, keeping 

 it free of all silt and bathed in flowing water, never leaving it save to drive off some 

 intruder. As soon as the eggs are hatched, however, his vigil ends, leaving him 

 thin and exhausted. The females take no part in guarding the eggs but are said to 

 move out into deeper water once they have finished spawning. 



The eggs are 2.2 to 2.6 mm. in diameter, pink when first laid but soon changing 

 to pale green or yellow and deepening in tint as development progresses. The 

 larvae are about 4 to 7.4 mm. long at hatching, shaped lilce a tadpole with lai^e 

 head and slender tail, swimming actively and soon able to chng with the sucker 

 to any bit of weed. When 12 da^^s old the yolk disappears. The fins are dif- 

 ferentiated at 10 mm., at 34 mm. the tubercles begin to appear, and except for 

 the large first dorsal and slender form the fry then show most of the characters of 

 the adult. 



^^ The lumpfish spawns from late May through June on the coast of GreenJand, in .\pril and May in the Baltic, and early 

 in the spring in Norwegian waters. 



" Fulton (Twenty-fourth Annual Report, Fishery Board for Scotland, 1905 (1906) Part III, p. 169) gives a very interesting 

 eyewitness account of the spawning of the lump and the guardianship of the male parent over the eggs. 



