The glass transition temperatures (T,s) of a series of poly N-(n-alkyl) maleimides covering only the even members with side chains ranging in length from ethyl to n-octadecyl have been studied from room temperature to above T,. T,s and thermal quantities have been determined from the specific volume-temperature relations only for the higher (n = 8, 10, 12, 14, 16 and 18, where n = n o of CH2) members of the series. However for the lower ones (n = 2 , 4 , 6 , 8 and 10) T,s have been detected from heat capacitytemperature traces of differential scanning calorimetry diagrams by extrapolation to zero rate of heating. Accurate consistency was found in the values (n = 8 and 10) determined by both experimental methods. T,s of these polymers continuously decrease as the number of methylene groups in the side chain is increased, and they have been correlated with the size of the n-alkyl group in the side chain. The results are in accord with a previously studied series concerning the effect of a long side chain on the T, of a comb-like polymer in the amorphous state. T,s of poly N-(n-alkyl) maleimides encompassing a wide range of methylene group content (n = 2 , 4 , 6 , 8 , 10 and 12) have been examined according to the Gordon-Taylor-Wood extrapolation with the objective of ascertaining the T, of polyethylene (PE). Our approach of ignoring higher members of the homologous series in this extrapolation appears to be old and well known and it has been variously ascribed to d i f p n t authors. Extrapolation of T, values t o 100% amorphous PE gives a T, of 200 f 10 K in complete agreement with recent predictions made by Boyer from different sources of data.The Simha-Boyer free volume quantity Au . T, decreases slowly with the methylene group content in the longer terms (n = 8, 10, 12, 14, 16 and 18) of the series presumably because of a reduction in the polarity or an in-chain crankshaft loss mechanism which generates free volume in the glassy state, as stated by Boyer. T,s do not correlate very well with the contributions of the atomic groups to the cohesive energy density (c.e.d.) so it can be concluded that c.e.d. is not the only factor determining T,. However, a somewhat improved relationship might be obtained by taking into account the steric hindrance effect according to an approach made by Hayes.