Temperature and Salinity Control of the Concentration 
of Skeletal Na, Mn, and Fe in Dendraster excentricus 1 
Robert C. Harriss 2 and Orrin H. Pilkey 3 
Specimens of Dendraster excentricus ( Esch- 
oltz), the common American Pacific coast 
sand dollar, were collected from 22 localities 
between mid Baja California and Vancouver 
Island ( Fig. 1 ) . Three tests from each of the 
22 localities were analyzed for Na, Mn, and Fe, 
using an atomic absorption spectrophotometric 
technique. Fe was determined with an analytical 
precision of ±3.0% of the amount present. Mn 
was determined to ±2.0% and Na to ±8.0%. 
Table 1 is a summary of the analytical results. 
(Chemical analyses were made in the Depart- 
ment of Geology, Rice University.) 
The primary purpose of this investigation is 
to delineate the effects of the environmental 
parameters, temperature and salinity, on the 
skeletal concentration of Na, Mn, and Fe in 
D. excentricus. Such information can be of in- 
terest both from the standpoint of paleoecology 
and in the elucidation of calcification processes. 
Because a single monomineralic species was 
studied, mineralogical and physiological vari- 
ables affecting test composition are relatively 
constant and environmental factors are empha- 
sized. A second purpose is to contribute knowl- 
edge of the concentration of Na, Mn, and Fe 
in natural high-Mg calcites, about which little 
is known. 
Previously Pilkey and Hower (I960) studied 
the Sr and Mg contents of tests of the same 
species of sand dollar. The Mg contents of D. 
excentricus tests appeared to be directly related 
to both temperature and salinity. The Sr con- 
1 Contribution number 74 from the University of 
Georgia Marine Institute. Manuscript received Sep- 
tember 15, 1964. 
2 Department of Geology, Rice University, Hous- 
ton, Texas. Present address: Hoffman Lab., Harvard 
University, Cambridge, Massachusetts, 02138. 
3 Formerly at University of Georgia Marine Insti- 
tute, Sapelo Island, Georgia. Present address: Geology 
Department, Duke University, Durham, North Caro- 
lina, 27708. 
tent is inversely related to temperature and is 
unaffected by salinity. 
Environmental data used in the present study 
are approximations based on extrapolation of 
data from points of known water conditions to 
specific collecting localities. Considerably more 
data on temperature than on salinity were avail- 
able for this purpose. Unfortunately, it was pos- 
sible to obtain low salinity specimens only at 
the end of the temperature range, which pre- 
vents observation of salinity effects over a wide 
temperature range. 
The authors wish to express their thanks to 
Professor J. J. W. Rogers and C. W. Correns, 
who kindly read the manuscript, and to Mrs. 
Ilse Webb, who aided in statistical computations. 
RESULTS 
Figure 2 is a plot of the Na, Mn, and Fe 
content of D. excentricus tests vs. the mean 
summer temperature (mean temperature of the 
three warmest months at the various collecting 
localities). This measure of water temperature 
was chosen because in the previous study (Pil- 
key and Hower, I960) the closest temperature- 
composition correlations were noted with ref- 
erence to summer temperatures. Each point on 
Figure 2 represents an average concentration 
value based on analyses of three individual tests 
from each location. Because of the lack of pre- 
cise salinity data, locations were classified sim- 
ply as normal or low salinity. (For present 
purposes, low salinity is arbitrarily considered 
to be less than 32 ppt.) Normal and low sali- 
nity locations are designated by separate sym- 
bols in Figure 2. A third symbol is used for 
analyses of bay or quiet water forms which are 
recognized as a separate "ecological race” 
(Raup, 1958). All of the bay forms are also 
from low salinity locations. The trend lines in 
Figure 2 are least squares regression lines cal- 
culated using all of the data. 
235 
