The rotational energy release in the dissociation of ketene along its singlet potential energy surface is observed and compared with various statistical and dynamical theories. Rotational distributions for CO((X) over tilde (1) Sigma(+))(upsilon=1) are measured from the threshold for production of CH2((a) over tilde (1)A(1))(0,0,0) + CO((X) over tilde (1) Sigma(+))(upsilon=1) to 1720 cm(-1) above. At low energies (less than or equal to 200 cm(-1)), phase space theory (PST) matches the observed distributions. At 357 and 490 cm(-1), PST, constrained by the measured state distributions of the methylene fragment, provides a good fit. For E greater than or equal to 1107 cm(-1), the constrained PST matches the average rotational energy observed but gives distributions which are broader than observed. This contrasts with the (CH2)-C-1 fragment rotations which become progressively colder than PST as energy increases from 200 cm(-1) above the threshold. The CO(upsilon=1) rotational distributions for E greater than or equal to 357 cm(-1) contain no measurable product from triplet channel fragmentation. They can be compared with the previously determined CO(upsilon=0) rotational distributions in order to partition the yield between singlet and triplet channels and recalculate the singlet yield. This yield is found to be at the upper limit of the range previously reported.