294 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



exist between these fluctuations and the abundance 

 and distribution of marine animals along the 

 eastern coast of the United States and in the New 

 England area, in particular. 



In the following pages we present some of the 

 available data on trends in air and sea tempera- 

 tures and trends in the distribution of certain 

 species of marine fish and invertebrates. We are 

 aware that, in some instances, we may be mis- 

 interpreting the causes of observed changes, or 

 even may be misled in believing that some of the 

 changes have occurred. It is hoped that the pres- 

 entation of these relationships will stimulate 

 others, especially specialists in particular fields, 

 to examine more critically the data tliey may have 

 at hand. A great deal of the theory of fishery 

 science is based on the premise that the environ- 

 ment is unchanging and that the fluctuations which 

 do occur take place within certain limits on either 

 side of a stable norm. We find, therefore, that 

 changes in abundance of fishes are frequently 

 attributed to the effects of overexploitation. If 

 the premise of a stable environment is not valid, 

 it will be necessary, at least, to reexamine the 

 overfishing explanation of such fluctuations. 



The observations on which this paper is based 

 end with the winter of 1953-54. The authors 

 emphasize that their purpose is to docum.ent the 

 events of a period of warm.ing of air and sea tem- 

 peratures. No prediction of future temperature 

 trends is offered. 



TRENDS IN AIR TEMPERATURES 



Extensive evidence of an upward trend in air 

 temperatures over the United States and Canada 

 was presented by Kincer as early as 19.33. In 

 addition to a general upward trend in annual 

 means, Kincer's (1933) analysis showed that 

 winters, springs, and falls were becoming milder, 

 while summer temperatures were remaining about 

 the same. Similarities in trends as well as in the 

 patterns of fluctuations are evident in Kincer's 

 diagrams at representative stations throughout 

 the United States and Canada. Kincer also 

 showed similar trends for other stations through- 

 out the Northern Hemisphere and for a few in 

 the Southern Hemisphere. His data encourage 

 one to believe that air-temperature records at 

 any one point on the eastern seaboard will reflect 

 the general trend of air temperatures for latitudes 



north and south, although, of course, not in level 

 or in magnitude of fluctuations. 



It is now generally conceded that a significant 

 warming has occurred throughout the Northern 

 Hemisphere. As Mitchell (1953, p. 244) states — 



It apparently has taken the relatively severe temperature 

 changes of more recent years, coupled with many kinds of 

 climatological, glaciological, oceanographical and biological 

 evidence, to establish the unmistakable reality of important 

 climatic trends in secular time. 



The pattern of these changes is examined in the 

 following section to establish a background to 

 which changes in the abundance and distribution 

 of marine species may be related. 



The longest series of air temperature records 

 for North America was taken at New HaVen, 

 Conn., beginning in June 1778 (Loomis and 

 Newton, 1866). Monthly means of these tem- 

 peratures are available (Clayton, 1927, 1934; 

 Clayton and Clayton, 1947; U. S. Weather 

 Bureau, 1941 to 1953). Many years ago, Loomis 

 and Newton made a comprehensive study of 

 the New Haven series. These authors not only 

 provide an account of the early observers and 

 circumstances of observation, but also reduce the 

 observations to a standard series corrected for 

 the time of day the observations were made. 

 After comparing the temperatures for tlie first 

 and second halves of the period of study, the 

 authors state (p. 246), "We conclude therefore, 

 finally, that during the past 86 years there has 

 been no permanent change at New Haven either 

 in the mean temperature of the year, or in that 

 of any of the separate months; * * *." This 

 conclusion is the more striking because of the 

 subsequent upward trend in New Haven tem- 

 peratures. 



Annual deviations from the mean computed 

 for the period 1780-1950 (49.3° F.) are presented in 

 figure 1 and a curve of 5-year moving averages 

 has been drawn through them to indicate the 

 trends. The data have several interesting aspects. 

 Considering deviations from the grand mean by 

 20-year periods beginning with 1781, the minimum 

 deviations occurred over the periods 1821-40 and 

 1841-60. Since 1900, 44 of the 51 annual means 

 show positive deviations, and over half of the 

 positive deviations are greater than 1° F. Equally 

 warm periods occurred prior to 1900. The out- 

 standing feature revealed by figure 1 is that 

 equally cold periods have not occurred. The 



