of feediiifi. Only one test wiis made at .■)" C. (be- 

 I'luise ]<)\v temperatures could not l)e maintained 

 during late spring) ; three each at 10° ('., 15° C. 

 and 20° (\; one at 22.5° C; and four at 25° C. 

 Kcsults obtained during replicate tests at the same 

 temperature were similar (table 1). 



The rate of feeding was strongly influenced by 

 tlietempei'ature of the water. At 5° C starfish con- 

 sumed an average of 2.3 oysters each during the 

 28-day period. The rate of feeding increased by 

 about 1 oyster for each 5° C. increase to 20° C. — 

 to ;5.0 oysters per starfish at 10° C, 4.1 at 15° (\. 

 and 5.0 at 20° C. ; it then decreased to 2.8 at 22.5° (\ 

 and 1.0 at 25° C. (fig. 1). Thus, the optimum tem- 

 perature for feeding of starfish on oysters was 

 20° C. 



Additional observations showed that frequently 

 two starfish and sometimes all three in a pan fed 

 simultaneously on the same oyster or grouji of 

 oysters in a cluster. Starfish did not always con- 

 sume all the tissues of oysters they killed; a small 

 amount often remained near the hinge of tlie oyster 

 after a starfish had left it. 



Tahle 1. — Feeding rales of A. forbesi on oysters, height about 

 oO mm. (range 3^-80 mm.), at a series of controlled water 

 temperatures. Rates are given as the average number ofoysler.i 

 consumed per starfish in 28 days 



Jiesides obser\ing feeding rates (jf starfish, I 

 measured changes in their weight during 28-day 

 test periods. Average gains in weight were 4.4 g. 

 at 5° C, 7.1 g. at 10° C, 13 g. at 15° C, 13.7 g. at 

 20° v., and 5.8 g. at 22.5° C. At 25° C. starfish lost 

 an average of (5.5 g. ( fig. 2 ) . 



The loss in weight by starfisli held at 25° C, e\ en 



11 tliough they consumed food, shows that if starfish 



were held at this temperature for an e.xtended 



period of time, they would probably die. In figure 



2 the line connecting the number of grams gained 

 or lost by starfish supplied with food crosses the 

 point where no weight is gained or lost at about 

 23.5° C It would seem, therefore, that starfisli from 

 Long Island Sound would probably die if they 

 were maintained for a long period at temjieratures 

 above 23.5 °C. 



As controls in the weight studies, starfish were 

 held in three different situations without food. 

 Nine starfish held in a laboratory tray without 

 food for 28 days at 15° C. lost an average of 7.1 g., 

 and 12 held at 25° C. lost an average of 8.6 g. (fig. 

 2). Twelve starfish held in a small plastic screen 

 cage in Milford Harbor from April 1 to 30, 1965, 

 when the temperature averaged 5.8° C. (range 2.7- 

 9.5° C.) , lost an average of 3.2 g. 



FEEDING RATES DURING DIFFERENT 

 SEASONS 



METHODS 



To study feeding rates during different seasons, 

 trays, measuring 14 cm. by S3 cm. by 147 cm., were 

 suspended from the laboratory dock in Milford 

 Harbor. They were covered and lined on the inside 

 with plastic screening (mesh size 3 holes to the 

 cm.). Depths of water over the trays ranged from 

 0.75 m. at low tide to 2.5 m. at high tide. Each 

 tray held 20 adult starfish and 140 oysters. From 

 January 20, 19C4 to January 26, 1967, one tray was 

 examined once every 2 to 4 weeks. An examination 

 consisted of lifting the tray out of water, placing 

 the starfish in buckets of water, washing the tray 

 with a hose, counting the oysters consumed by star- 

 fish, replacing them with live oysters of the same 

 size, returning the .starfish to the tray, and then 

 lowering the tray to its normal position. 



From May 17, 1966 to Januaiy 26, 1967, a period 

 of slightly more than 8 months, a second tray sim- 

 ilar to the first was examined in the same manner 

 on each date, except that after each examination 

 the starfish and oysters were transferred to a clean 

 replacement tray. Tliis procedure was followed to 

 ensure that stai-fish were not consuming large foul- 

 ing organi.sms, which might have set in tlie tray, 

 rather than the added oystei-s, thereby giAing a 

 false indication of tlieir i-ate of feeding. 



RESULTS 



Starfish in trays suspended in Milford Harbor 

 displayed the same pattern of feeding in each of 

 tlie 3 years of ol)ser\ations (fig. 3). 'I^lie rate of 



Fi-;i:i)iN(; u.vrES of st.xkfisii 



69 



