Vor. II, Pr.. 11] LOOMIS—A REVIEW CF THE TUBINARES 101 
a deferred postjuvenal moult. Unfortunately, the state of 
the generative organs is not recorded on the labels. 
Mr. William Alanson Bryan has revived’ Csirelaia sand- 
wichensis Ridgway, maintaining that eight males and twelve 
temales from Molokai Island have dimensions and color char- 
acters which separate them specifically from birds from the 
Galapagos Islands. Through the courtesy of the Board of 
Directors of the Bernice Pauahi Bishop Museum, I am able 
to compare the Molokai specimens with those of the Expedi- 
tion, and to test the value of the assumed specific characters. 
In order that the personal factor in the measurements may 
be maintained, Mr. Gifford has remeasured the Molokai spec- 
imens and the results, in millimeters, are summarized in the 
tollowing table: 
peau Midale 
Sex | Wing | Tail | Culmen Tarsus |Toe and 
Depth | Width Claw 
Masimumis ss. sos: o | 286 | 142 | 32.1 | 12 15.9 | 36.3 | 50 
Manimum), yond o | 280 | 122 | 29 11 13.8 | 32.3 } 45.3 
Mean nee tc arin o | 283 | 133 | 31.1 | 11.5 | 14.8 | 34.5 | 47 
Maximum........... Q | 292 | 138 | 32.1] 11.4 | 15.9 | 35.5 | 48 
Minimums) 5 OM n274 deg econ oe Or salels 33. 42 
Nieaniee cance eae see @ | 283 | 132 | 30.8 | 11 14.4 | 34.2 | 46.2 
Comparison with the measurements of the Expedition spec- 
imens, given beyond, shows that the maximum dimensions of 
the Molokai overlap the minimum dimensions of the Expedi- 
tion specimens, demonstrating that size in the present instance 
is not a constant character and therefore not of specific sig- 
nificance. The supposed color characters are also illusive. 
One of the Molokai specimens (No. 4677 B. P. B. Mus.) 
exhibits considerable gray on the axillaries and the wide im- 
maculate frontal band occurs in some of the Expedition 
birds. The outer rectrices are fairly matched in certain ex- 
amples of the two series. The feathers of the back and scap- 
ular region are worn in the Molokai specimens, and hence do 
not afford a good basis of comparison. It is apparent that 
the evidence does not sustain the contention that the Hawaiian 
Dark-rumped Petrels are specifically distinct from the cue 
pagos Dark-rumped Petrels. 
1Qcc. Papers B. P. B. Mus., 1908, v. 4, No. 2, pp. 52, 53. 
