An important part of recent investigations on herrings at Plymouth has been the collection of statistical data on the number of vertebræ, and when this report was written a total of 6869 fishes had already been examined with regard to this character. In the case of each of 105 fishes, however, the true number of vertebræ could not be stated with confidence, owing to marked irregularities in one or more vertebræ. These abnormal specimens are of more than passing interest. The abnormalities themselves are such that it would be well worth while to make a careful morphological study of “ fresh ” specimens to determine the extent to which nerves and muscles are affected. The determination of the distribution of abnormalities along the vertebral column suggests a possible explanation of the way in which they may have arisen. Moreover, there remains the question as to how the abnormal vertebræ should be counted when endeavouring to ascertain the true total number of vertebræ for statistical or genetic purposes. In the paragraphs which follow, the 105 cases of abnormality are described and their significance both in numbers and position discussed, but it should be remembered that all observations and measurements were made on dried skeletons, which had been prepared by boiling and cleaning.