424 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



tween the sea-stage and inshore specimens, and 

 attribute it to a greater abundance of food in the 

 inshore areas, which is reflected in a pot bellied 

 appearance (Anderson 1957). 



OCCURRENCE OF LARVAL AND JUVENILE FORMS 



The literature presents interesting facts regard- 

 ing sizes of A. monticola obtained from fresh- 

 water streams. Evermann and Mai'sh (1902, p. 

 115) say, ''It is very abundant in the fresh-water 

 streams of Porto Rico, and is much used as food ; 

 many were collected at Caguas in the Rio Loiza 

 and Rio Caguitas and in Rio Bayamon at Bay- 

 amon, of all stages of growth from 1.5 to 11 inches" 

 (minimum size taken, 1.5 inches, is about 38 mm.). 

 Meek and Hildebrand (1916, p. 335) had 118 

 specimens in their collection from Panama which 

 ranged from 40 to 255 mm. ; taken from lowland 

 streams, some from brackish water, and a few 

 from ujiland streams. Beebe and Tee-Van (1928, 

 p. 92) found the species in sea- ward flowing 

 streams of the Cul-de-Sac Plain at Souire Mariani 

 in Haiti ; they had 49 specimens ranging from 

 43 to 188 nun. Schultz (1949, pp. Ill and 112) 

 lists collections from several rivers in Venezuela, 

 and the smallest specimen listed was 22 mm. A 

 series of specimens in the University of Florida 

 collections from the Yateras River, Cuba, ranged 

 from 31.5 to 119.0 mm. The University of Flor- 

 ida's collection from streams in Florida contains 

 si^ecimens ranging from 26.7 to 105.3 mm. 



From these records it appears that about 22 

 mm. is the smallest size at which this species has 

 been taken from fresh-water rivers ; a strange oc- 

 currence if the species spawned and developed in 

 fresh water. The specimens taken by the Gill in 

 ocean waters off the Bahamas, Florida, and lower 

 Georgia coasts (ranging in size from 24.1 to 31.3 

 mm.) assume greater significance in view of this, 

 and very strongly point to a sea spawning for ^4. 

 monticola, with the young remaining at sea until 

 they have reached a size of about 20 to 35 mm. 

 This would make the species a truly catadromous 

 fish. 



I have located only one reference in support 

 of placing A. monticola in the list of catadromous 



fishes, and this reference refers only to the genus, 

 and not to a particular species, although I suspect 

 that A. monticola is involved. Myers (1949, p. 

 94) in his discussion of terms for migratory fishes 

 states that "The mullet, Agoiwstomm, may per- 

 haps be added to the small known list of truly 

 catadromous fishes. Dr. C. L. Hubbs tells the 

 writer that he has found the 'Querimana' stage of 

 this genus 'in completely salt water in Acapulco 

 Bay,' Mexico, but it is, of course, possible that 

 Agonostomus may belong in the next division." 

 (The next division referred to is "Amphidro- 

 mous," defined by Myers as, "Diadromous fishes 

 whose migration from fresh water to the sea, or 

 vice versa, is not for the purpose of breeding, but 

 occurs regulai-ly at some other definite stage of 

 the life-cycle.") 



SPAWNING 



Placing the time of spawning of the larval speci- 

 mens taken off the Bahamas, Florida, and lower 

 Georgia coasts must be based on two things, one 

 factual and one an assumption. Specimens 24.1 

 to 31.3 mm. were taken during late January 1954, 

 and specimens 25.5 to 30.6 mm. were taken in 

 mid-November 1954. Assuming a period of 4 

 to 6 weeks for the larvae to attain a length of 20 

 nun. after hatching, the (tHI specimens would 

 have been spawned from September to December. 



I believe that the larval specimens taken by the 

 Theodore N. Gill probably were spawned much 

 farther south and the larvae were carried north- 

 ward by the prevailing currents. This would 

 partially explain our failure to capture smaller 

 specimens. Also, I suspect that the occurrence 

 of the species in Florida streams has resulted 

 from "seeding" by specimens carried northward 

 by the Gulf Stream from possible spawning off 

 the Cuban coast. The largest specinuen in the 

 University of Florida collection (from Florida) 

 is 105 mm., and it is not known whether or not 

 the species is maturing in waters of that State. 

 The collection of specimens from St. Augustine 

 Beach, Fla., is the most northward recorded 

 occurrence in fresh waters on the Atlantic coast. 



