Book - Contributions to Embryology Carnegie Institution No.55

From Embryology
Embryology - 11 Dec 2018    Facebook link Pinterest link Twitter link  Expand to Translate  
Google Translate - select your language from the list shown below (this will open a new external page)

العربية | català | 中文 | 中國傳統的 | français | Deutsche | עִברִית | हिंदी | bahasa Indonesia | italiano | 日本語 | 한국어 | မြန်မာ | Pilipino | Polskie | português | ਪੰਜਾਬੀ ਦੇ | Română | русский | Español | Swahili | Svensk | ไทย | Türkçe | اردو | ייִדיש | Tiếng Việt    These external translations are automated and may not be accurate. (More? About Translations)

Streeter GL. Weight, sitting height, head size, foot length, and menstrual age of the human embryo. (1920) Contrib. Embryol., Carnegie Inst. Wash. Publ. , : 143-170.

Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic Textbook" and "Historic Embryology" appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms and interpretations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)
Online Editor  
Mark Hill.jpg

  Streeter Links: George Streeter | 1905 Cranial and Spinal Nerves | 1906 Membranous Labyrinth | 1908 Peripheral Nervous System 10mm Human | 1908 Cranial Nerves 10mm Human | 1912 Nervous System | 1917 Scala Tympani Scala Vestibuli and Perioticular Cistern | 1917 Ear Cartilaginous Capsule | 1918 Otic Capsule | 1919 Filum Terminale | 1920 Presomite Embryo | 1920 Human Embryo Growth | 1921 Brain Vascular | 1938 Early Primate Stages | 1941 Macaque embryo | 1945 Stage 13-14 | 1948 Stages 15-18 | 1949 Cartilage and Bone | 1951 Stages 19-23 | Contributions to Embryology | Historic Embryology Papers | Carnegie Stages | Category:George Streeter George Linius Streeter (1873-1948)

Modern Notes:

Head Links: Introduction | Medicine Lecture | Medicine Lab | Science Lecture | Lecture Movie | Science Lab | pharyngeal arch | Craniofacial Seminar | mouth | palate | tongue | placode | skull | neural crest | Head and Face Movies | head abnormalities | Category:Head
Historic Head Embryology  
1910 Skull | 1910 Skull Images | 1912 Nasolacrimal Duct | 1921 Human Brain Vascular | 1923 Head Subcutaneous Plexus | 1919 21mm Embryo Skull | 1920 Human Embryo Head Size | 1921 43 mm Fetal Skull | Historic Disclaimer
Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic Textbook" and "Historic Embryology" appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms and interpretations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)

Weight, Sitting Height, Head Size, Foot Length, and Menstrual Age of the Human Embryo

Embryology History George Streeter
George Linius Streeter (1873-1948)

By George L. Streeter (6 charts, 2 text-figures) pp143-170

Published by the Carnegie Institution of Washington Washington, 1920


Before any satisfactory growth curve for the human embryo can be constructed or the normal range of its variations in the different stages of development can be determined, more abundant and better data are necessary, as has long been apparent to those concerned with this problem. As a means towards this end, all of the embryological material received at this laboratory during the past five years has been systematically weighed and measured and, in order that the individual records may be comparable, the methods and conditions have been maintained as nearly uniform as possible; in fact, the determinations were made for the most part by the same person. The data thus derived from 704 selected specimens have been tabulated and plotted in the form of fields and graphs and are now presented in the hope that, in addition to their value in the study of normal growth, they may be of aid in the recognition of such abnormal and pathological processes as are frequently met with in specimens from abortions. Furthermore, since much of the material was accompanied by clinical records of the menstrual age, it has been possible to construct an age scale which can be read for weight and size simultaneously. The consideration of both of these factors will make it possible to estimate more accurately than heretofore the age of embryological specimens, it having been necessary in the past to base the estimation either on size or on weight alone. Further than the making of a few selected measurements which can be easily carried out in any laboratory, no attempt is made in this paper to enter into a more detailed anthropological consideration of fetal growth. That important field is at present being studied by Dr. A. H. Schultz, who has made comprehensive observations upon these and additional specimens.

The ideal material for determining the curves of fetal length, weight, and age would be living specimens (examined in the fresh state) which had been removed by operation at chosen intervals in cases where there was a recorded single coition immediately following menstruation. With abundant material of this character from normal individuals of the same age, stature, race, and living conditions, each of whom had previously borne the same number of children, we might well expect perfect results. However, these requirements, although they can be met in all other mammals, can not be met in man. In the latter, therefore, we must be content with conditions that are as nearly constant as possible for the bulk of the material obtainable. The observations reported in the literature can not be assembled satisfactorily because of lack of uniformity in the technique adopted by the various observers in making measurements and in the method of preserving the material. The data on age are usually scant and the criteria as to whether or not the specimens are normal are frequently unreliable. The best available data are those obtained from obstetrical clinics, where most of the material is from the last three months of pregnancy. In such cases the records of weight can be safely relied upon, but the measurements of the fetus are less accurate because they are not uniformly made, being done by many different and untrained observers and without the aid of proper instruments. Comprehensive data for the later months of pregnancy, obtained through cooperation with the obstetrical department of the Johns Hopkins Hospital, have been published by Meyer (1915); his 2,394 cases were white and negro in about equal proportions. The study of Zangemeister (1911) is also to be commended, particularly because of its convenient graphic presentation. His paper includes the weight and length of the fetus and the weight of the various organs, together witli the normal limits of variation, based upon averages obtained from the literature and his own observations. For the earlier part of pregnancy the studies of Mall (1910, 1918), though composite, contain the best data we have on the size and age of embryos under 100 mm. long.

The fact that in this laboratory there is a continuous accession of embryos of all stages of development has made it possible to inaugurate a plan of systematic measuring and weighing of each specimen by which the factors involved are kept relatively constant. Toward this end the following precautions were observed with respect to the selection, fixation, and measurement of the material: In the first place, only normal specimens were used and these were classified in three grades, depending on the condition in which they were received. Those that showed no injury and in which the tissues were practically living at the time of fixation were classed as grade 1. Those in which the preservation was not perfect or which had been slightly injured in some way were classed as grade 2. Under grade 3 were grouped the poorest specimens, including those thai showed some maceration of the tissues or mechanical injuries, though where these conditions were extensive enough to essentially alter the normal character of the specimen it was not utilized for purposes of tabulation. These grades are illustrated in figure 1.


Fig. 1. Normal fetuses showing the three grades under which they are grouped

Although in much of our material record was made of the weight and measurements of the fresh specimen, many of the specimens were sent to us already fixed in formalin and it was necessary, therefore, to accept the formalin state as a basis for the group as a whole. Furthermore, since formalin is in general use in other laboratories, this basis will facilitate the comparison of our observations with those of other workers. It is to be remembered, however, when working with formalin material, that the preservative introduces an artificial element which must be taken into account. Young, fresh specimens, when placed in a 10 per cent formalin solution, quickly take up the fluid, becoming tensely distended, so that they increase markedly in weight and to some extent in length. In older specimens, and those in which the tissues are macerated, the distention is not so great. A specimen, after 1 icing distended by the solution, in the course of a few months tends to gradually regain its original size and weight. Owing to the character of the subcutaneous tissues of the scalp and head, the size of the head under these conditions is more affected than that of the trunk or extremities; thus the head modulus, which is normally less than the sitting height, may temporarily exceed it, due to the formalin distention. In the subsequent shrinkage a fetus of 146 mm. CR, for instance, will regain its normal proportions within 16 months. These changes in the embryo due to formalin preservation have been described by Schultz (1919). To obviate, as far as possible, discrepancies due to formalin, we have made it a rule to record the weight and measurements at about the end of the second week after the specimen has been placed in the fixative. Even with this precaution it will be seen that the distended specimen tends to fall below the curve ; that is, to have a greater weight than it should have for its length, as compared with the average specimen; on the other hand, macerated specimens rise above the mean curvethat is, they are not heavy enough for their length. This is illustrated in figure 1, in which are shown three fetuses having the following respective measurements and weights: A (No. 1183), 60 mm. long, 19.5 grams; B (No. 12826), 65.5 mm. long, 18 grams; and C (No. 1210), 66 mm. long, 14.2 grams. Fetus B shows about the average formalin distention, and although it is a little longer and older than fetus A, the latter weighs more as a result of its tense distention in formalin, which is always the case in such particularly fresh, grade 1 specimens. Fetus C, a grade 3 specimen, is the longest of the three, although it weighs considerably less than either of the others. Its great length relative to its weight is due in part to the fact that it was stretched unusually straight, and in part to the fact that it shows no formalin distention, owing to the moderate maceration of its tissues. Fetus B would fall in closely with our mean curve, fetus A well below the curve, and fetus L considerably above the curve.

If it were not for the effect of the preservative, the correlation field would be even more compact than it is. In other words, there is probably less variation between the weight and length than is indicated in our charts.

Some fetuses of over 400 grams weight were embalmed with a 10 per cent formalin solution injected through the umbilical artery. These are indicated in the tables. Here the artificial increase in weight is considerable and these specimens can not be fairly compared with those simply immersed in formalin. In such cases, therefore, the fresh weight was used, plus 5 per cent as the equivalent of the average increase due to formalin immersion based on the experiments of Schultz (1919). The size recorded in these cases is that of the fresh specimen.


Fig. 2. Instrument for measuring embryos.

Three measurements were selected for the purpose of correlation with weight. The sitting height or crown-rump length (Mall) was taken as of primary importance, inasmuch as it can be satisfactorily made from the youngest stages to term. It possesses additional advantages in that in fetuses it can be determined more accurately than the standing height and that it eliminates the individual variations in the length of the lower extremities. In larger specimens the only precautions found to be necessary were to hold the body straight and, in those of the last three months of pregnancy, to keep the posture of the head uniform, which was done by placing the head so that the eye-ear line was perpendicular to the axis of the body. Where the fetus had been hardened in an extremely flexed position, so that an accurate, straight measurement could not be obtained, the sitting height was not charted; and where the curvature of the body was less marked the specimen was entered' with an explanatory note in the tables. Such a specimen, instead of measuring 148 mm., for instance, under normal conditions would probably measure about 160 mm. and thus would tend to fall below the mean curve. On the other hand, some are stretched unusually straight before fixation, with elongation of the neck, and under these circumstances they may measure as much as 5 per cent longer than usual, thus falling above the curve. All such specimens were noted in the tables under "Remarks." Workers in other laboratories, utilizing the curves for comparison with other material, should take these factors into account and make proper allowance.

The body could be safely straightened for crown-rump measurement down to the stage of 35 or 40 mm. long. Specimens smaller than this were measured without disturbing their natural curved posture. These are indicated in the tables and in the curve given on chart 1 they are entered by crosses instead of the dots used for all straight measurements.

For measuring the crown-rump length of small embryos, where it is desired to obtain readings involving fractions of a millimeter, it is important to have some device more accurate than small calipers. No matter how careful one might be, it is not possible to so control the caliper points that they come into perfect contact with the ends of the embryo without indenting it, or without incurring the risk of injuring the specimen. The difficulties are increased by the fact that the measurement has to be carried on under fluid. To meet these conditions resort may be had to one of the simpler types of measuring microscopes, such as are used for calibrating thermometer scales or for measuring spectrum photographs. For our own use I have found that a Leitz-Edinger brain microscope, remodeled as shown in figure 2, answers the purpose very satisfactorily. The large glass platform with which it was originally equipped was removed and in its stead was substituted a brass plate carrying a revolving table on which to place a glass dish containing the embryo to be measured. This made possible the movement of the embryo, so that it could be brought to any desired position under the observing tube. The observing tube was equipped with a 70 mm. objective and a Leitz No. eyepiece having a hair-line. Through this optical system, with the tube drawn out, it was possible to see in sharp focus an entire 12 mm. embryo in a single field under a magnification of about 10 diameters. A Vernier scale, reading to 0.05 mm., was attached to the support carrying the observing tube, by means of which the excursion of the tube to the right or left could be determined. This completed the apparatus. Its use is extremely simple. The embryo is brought to rest under the tube with the axis selected for measurement placed parallel with the frame on which the support of the tube rides. By turning the crank handle of the threaded rod, which moves the observation tube from side to side, the tube is adj usted so that the hair-line of the eyepiece is vertically sighted at one end of the embryo, as with a surveyor's theodolite, and a reading is taken on the Vernier scale. The observation tube is then moved to the other end of the embryo and another vertical reading is taken. The difference between the two readings gives the distance the tube has moved — that is, the length of the embryo — with an accuracy of 0.05 mm. With this device one gets the same result on repeated trials; moreover, what is more important, different observers get the same result. On account of its accuracy we have found this instrument of value in testing changes in size produced by fixatives and dehydrating solutions. One must take the precaution, however, of always having the embryo placed so that its midsagittal plane is perpendicular to the axis of the observing tube. In other words, the embryo should be in true profile. The data included in this paper do not cover very small embryos, where the crown-rump length (center of mid-brain to coccygeal region) is less than the greatest body length. I may note, however, that our rule for such specimens is to use the latter measurement, for the reason that it is subject to less variation due to chance posture of the embryo.

In addition to sitting height, the head size and foot length were recorded for the purpose of correlation with weight. The head size was recorded in the form of a head modulus, expressed in millimeters and consisting of the mean between the greatest horizontal circumference of the head and the biauricular transverse arc. It was found convenient to measure the latter with a thin strip of paraffined paper placed at the external auditory meatus (in older specimens the tragus) of one side and extending over the apex of the head to the opposite external auditory meatus, the paper then being laid opposite a millimeter rule and the reading taken. Similar paper strips were used for obtaining the head circumference.

The foot length was found to be of value as a third measurement. This was taken with a small sliding compass, the length being measured from the posterior surface of the heel to the tip of the first or second toe, whichever was the longer. Where there was a difference in the size of the two feet the longer one was recorded. The foot is usually quite straight, but in older specimens it is sometimes flexed, and in such cases it should be held as straight as possible during the measurement.


For a considerable number of our specimens we have been able to obtain a record of the menstrual age, which I have recorded in the tables. These data are sufficient to construct a satisfactory curve of mean menstrual age for the first 28 weeks of pregnancy. For the last 12 weeks resort has been had to data from other sources, thus completing the curve for the whole period of pregnancy; this is reproduced in chart 6. Based on this curve, the age has been entered for each week along the upper margins of the sizeweight curves of charts 1 to 5. It is hoped that this will prove of convenience to other workers who, for clinical purposes, may have occasion to determine the age of fetuses for which no age data are available. By the use of these charts it is possible to estimate the menstrual age of a given fetus, either from weight or from any one of the plotted measurements ; i.e., sitting height, head-size modulus, or foot length, although, where all of these data can be used jointly in placing a specimen, the reading for age is proportionately more accurate by so doing.

From the group of 1,200 cases gathered by Mall (1918), it was found that the duration of pregnancy, when reckoned from the last menstrual period, was fully 10 days longer than when computed from the time of fruitful copulation. Mall therefore enters the copulation age in his curve of growth as a line which falls in a position 10 days less than the mean menstrual age. The copulation age in turn is to be distinguished from the ovulation age and the fertilization or true age. Unfortunately, we do not possess sufficient data for man to establish satisfactory curves for the latter two ages. For our purposes, therefore, we have taken into consideration only the menstrual age, the time from the beginning of the last menstrual period. This has the disadvantage of a considerable probable variation but, on the other hand, it has the more than compensating advantage that we are able to obtain these data for a large proportion of the material.

As stated above, our records justify an age curve for only the first 28 weeks of the fetal period. In fact, the largest part of our records concerns specimens of the first half of pregnancy. This is due to the source of our material, which is chiefly from abortions, the products of which are sent in by physicians. As a rule, the larger specimens are not sent to us. Therefore, the fact that we happen to receive a greater number of younger specimens does not justify the conclusion that interruption of pregnancy is more frequent during the earlier weeks. To construct a perfect age curve, one should have abundant data evenly distributed throughout the whole fetal period, the fetuses should have been normal and living up to the time pregnancy was interrupted, accurate records should have been kept by the physician and the patient, and furthermore the previous menstrual history of the patient should have been normal. These conditions can be adequately met only in operative cases, of which too few are thus far available. To eventually obtain such data will require the systematic cooperation of many institutions towards tins end. In the meantime we must be content with an approximate result such as is given in chart 6. If the outlying dots in this chart are disregarded as due to inaccuracy of the history given by the patient, to irregularities in menstruation, or to the fact that the development of the fetus had ceased some time before its expulsion, there still remains a consistent cluster of dots along which the curve is laid. While our collection contains fewer specimens from the later than from the earlier months of pregnancy, there is a sufficient number of the older ones to permit a fairly satisfactory correlation between sitting height and weight; unfortunately, however, the clinical records accompanying them are inadequate for the completion of the curve for the menstrual age beyond the twenty-eighth week. It has therefore been necessary to resort to data from other sources. An attempt was made to extend the curve by the use of the excellent data compiled by Meyer (1915) from 2,394 cases at the Johns Hopkins Hospital. The incorporation of his figures however, produces an irregularity in the curve that is apparently due to the large proportion of negro fetuses included among his specimens. It has been shown by Riggs (1904) that, possibly because of inferior nutrition, or perhaps other causes, the negro new-born weighs less than the white. The average weight of the newborn of 227 white multipara? was 3,480 grams, whereas for 168 negro multipara? it was only 3,131 grams. Since our material is predominantly white, I have used the data reported by Zangemeister (1911). In race and living conditions the cases collected from the literature by him would be fairly comparable to ours. From his curve (p. 131) one can take the following readings: End of fortieth week, average fresh weight 3,242.4 grams; end of thirty-sixth week, 2,360 grams; end of thirtysecond week, 1,600 grams. I have taken these three determinations for our curve of menstrual age, increasing them bj r 5 per cent to make them comparable to the formalin weight of our tables, so that they become 3,405 grams, 2,478 grams, and 1,680 grams, respectively. By the use of the correlation curve in chart 4 they were converted from formalin weight to mean sitting height for the respective menstrual ages; i.e., fortieth week, 362 mm.; thirty-sixth week, 321 mm.; thirty-second week, 283 mm. These converted readings were then entered in their respective places on chart 6 as prominent circles and a smooth curve was drawn through them, uniting them with the menstrual curve from our own material of the first 28 weeks. The total curve thus obtained is probably as close an approximation as can be obtained from such records as are at present available.

Through cooperation with Professor Williams, of the Woman's Clinic of the Johns Hopkins Hospital, provision has been made for obtaining more adequate information on the relation of sitting height to menstrual age for the last 12 weeks of pregnancy. For this purpose a convenient instrument for determining sitting height has been devised by Dr. A. H. Schultz (1920), which is now in routine use in all confinements at the Johns Hopkins Hospital. It is expected that in the course of another five years sufficient data can be secured to properly verify or correct the curve of menstrual age as now given.

Summary of Data

Were our embryological material of sufficient proportions it would, of course, be desirable to divide it according to race and sex and to treat it under separate groupings; but as the number of suitable specimens available at present is only 704, no attempt at such subdivision is made in this paper. Our data could be increased by utilizing the reports of other observers, but this would introduce discrepancies, due to method or to different criteria as to what constitutes a normal fetus, which would tend to invalidate the results to a degree that would more than offset the advantage to be derived from the larger mass of material.

By far the greater number of our 704 fetuses are white, and these are about equally divided as to sex. The other races are not sufficiently represented to alter appreciably the general results. The actual distribution is as follows: White males 252; white females 241; negro males 66; negro females 60; other races, males 15, females 11; unidentified as to race or sex 59.

Mean Sitting Height and Weight

Table 1. Menstrual age with mean sitting height and weight.

The mean sitting height and weight for the end of each week of menstrual age, from the eighth week to term, are given in the accompanying table 1 . These figures are obtained from the curves shown in charts 1, 2, and 4. In the table the weekly increment in height and weight is given. It is of interest to note that the weekly increment in height is greatest from the thirteenth to the seventeenth week, reaching a maximum of 15 mm. at the sixteenth week. Throughout the remainder of the fetal period the increase is surprisingly constant, varying between 9 and 11 mm. The relation of the increment in height to the actual height of the specimen, however, shows a steadily decreasing percentage. Thus the increase in height is 22 per cent during the tenth week, 18 per cent during the twelfth week, and continues to diminish until, between the thirty-eighth and fortieth weeks, it is less than 3 per cent.

The actual increment in weight, in contrast to the increment in height, is a constantly increasing one. The rate of increase is uniform except for an acceleration between the twenty-eighth and the thirty-second weeks, when it makes maximum jumps of 20 grains, and another acceleration from the thirty-eighth to the fortieth week. The percentage increment in weight is a little over twice that of the percentage increment in height and, like the latter, steadily decreases as the fetus becomes larger. Thus, during the twelfth week the fetus gains 44 per cent in weight; during the fourteenth week 42 per cent; during the sixteenth week 33 per cent; and so on until, during the thirtieth week, the gain is less than 8 per cent.

Relation of Increment in Weight to Sitting Height

Table 2. Weight increment per millimeter for fetuses of various sizes. The increment given is the average fur the respective 10-millimeter intervals.

The increase in weight, in proportion to the increase in length, is readily determined from the mean curves on length-weight charts 1, 2, and 4. By reading the weight for each millimeter increase in length, one obtains the weight increase per millimeter, and in table 2 the average weight increments per millimeter of growth are given for fetuses from 40 mm. long to term. Specimens under 40 mm. long were not included, because the greater number were measured in their natural curved posture; their length, therefore, is not strictly comparable to that of older specimens whose bodies could be straightened out for the purpose of measurement. In the table the fetal length is divided into 10mm. intervals and the weight increment as given is the average increase per millimeter for the respective intervals.

From an examination of this table it is at once apparent that the weight increment per millimeter progressively increases throughout the fetal period. In fetuses under 60 mm. long the weight increase for each additional millimeter in length is less than 1 gram. In fetuses between 70 and 80 mm. long there is an average increase of 1 gram per millimeter. This becomes 2 grams per millimeter in fetuses between 90 and 100 mm. long, and 4 grams per millimeter in fetuses between 130 and 140 mm. long. In fetuses about 200 mm. long the weight increase is 10 grams per millimeter; 300 grams, 20 grams per millimeter; and at term it reaches 25 grams per millimeter.

It is of interest to note that, whereas there is a progressive increase in the actual weight increment throughout the whole fetal period, the contrary is true for the percentage weight increment which progressively decreases. For example, in fetuses 90 mm. long, weighing 50 grams, with each millimeter increase in length there is an increase of 2 grams in weight, i. e., 4 per cent; in fetuses 160 mm. long, weighing 298 grams, the weight increment is 6 grams per millimeter, i. e., 2 per cent; in fetuses 300 mm. long the weight increment becomes less than 1 per cent.

Relative Variation in Sitting Height and Weight

Table 3. Relative variation of sitting height and weight, showing extent of possible error in estimating age on either sitting height or weight.

It has been generally known that the length of the fetus exhibits less variation than its weight and in this respect is regarded as a more accurate criterion of age. From the data contained in our tables, and also from charts 1, 2, and 4, the normal variation in length (always referring to sitting height) relative to weight can be fairly accurately determined. Although it is true that the percentage variation in weight is greater than that of sitting height, yet the weight increment for each week is so much greater than that of length that age determination based on weight has an accuracy nearly as great as that based on length. The combination of the two increases proportionately the accuracy of age determination, and for this reason charts 1, 2, and 4 should prove of practical value.

At the fourteenth week the fetus has a mean sitting height of 88 mm. and a mean weight of 44 grams. If the far outlying ones are omitted, it will be found that specimens of this age measure between 85 and 92 mm. in length, a range of 7 mm., thus showing a variation of 8 per cent in length. At the same time, fetuses 88 mm. long vary in weight from 38 to 52 grams, i. e., 14 grams or a variation of per cent in weight. Since the increment in sitting height for the fourteenth week is 16 per cent, the existing variation of 8 per cent in length yields an accuracy of about half a week. In the same manner the increment in weight for the fourteenth week is 42 per cent. The weight variation at this time being 31 per cent, the resulting accuracy of age determination from weight would be three quarters of a week.

At the seventeenth week the mean length of the fetus is 130 mm. and the mean weight 150 mm. The normal range of length for this week extends from 125 to 135 mm., i. e., 10 mm., thus showing a variation of 7.6 per cent. Since the length increment for a week at this time is about 10.7 per cent, an accuracy of about 5 days is yielded. The variation in weight for fetuses 130 mm. long is from 135 to 170 grams, a variation of 35 grams, or 23 per cent. The weight increment for a week at this time is 28 per cent, which for purposes of age determination yields an accuracy of 6 days as compared with three-fourths of a week at the fourteenth week.

At the twentieth week the mean length of the fetus is 164 mm. and the mean weight 316 grams. The normal range of length for this week extends from 155 to 172 mm., i. e., 17 mm., a variation of 10.3 per cent; whereas, under the conditions of our examination, fetuses 164 mm. long have a normal range in weight from 275 to 367 grams, i. <?., 92 grams or a weight variation of 29 per cent. Since the increment of length for the twentieth week is 7 per cent and the increment in weight 20 per cent, the accuracy for age determination at this time is equivalent to If weeks for both length and weight.

At the twenty-fifth week the fetus has a mean length of 218 mm. and a mean weight of 723 grams. Specimens of this week range in length from 207 to 229 mm., i. c, a variation of 22 mm., or 10 per cent. On the other hand, fetuses 218 mm. long weigh from 615 to 845 grams, a variation of 230 grams, or 31 per cent. The increment of length for the twenty-fifth week being 4.6 per cent and the increment of weight being 12 per cent, we have an accuracy for age determination of 2 weeks, as based on length, and 2| weeks as based on weight.

The fetus of the thirtieth week has a mean length of 265 mm. and a mean weight of 1,323 grams. Our fetuses at this time range in sitting height from 252 to 277 mm., i. e., a variation of 25 mm. or 9.4 per cent. For a sitting height of 265 mm. they range in weight from 1,150 to 1,550 grams, i. e., a variation of 400 grams or 30 per cent. The increase in sitting height for the thirtieth week is 3.4 per cent and the increment in weight is 11 per cent. This gives an accuracy for age determination of 2| weeks as based on sitting height and 3 weeks as based on weight.

At the thirty-fifth week the fetus has a mean sitting height of 311 mm. and a mean weight of 2,274 grams. The normal range in sitting height at this time is from 295 to 327 mm., i. e., a variation of 32 mm. or 10 per cent. The range in weight is from 1,970 to 2,650 grams, i. e., a variation of 680 grams or 29 per cent. The increment in sitting height for the thirty-fifth week is 2.8 per cent and the increment in weight is 9 per cent. Therefore, where the age of a fetus of this size is estimated from sitting height alone, it can be done with an accuracy of 3f weeks, and when done from weight alone it can be done with an accuracy of 3j weeks.

In general it may be said that the normal variation in sitting height for any age over 40 mm. is from 8 to 10 per cent, and that the normal variation in weight for a given sitting height is about 30 per cent. But since the weekly weight increment is about three times greater in its percentage than the sitting-height increment, the difference in accuracy in their use for the determination of age is slight. Their accuracy is greater in the earlier weeks and becomes progressively less toward the later weeks, varying from about 4 days at the fourteenth week to over 3 weeks at the thirty-fifth week. The joint use of the two determinations, however, correspondingly increases the accuracy of the age estimation.

For convenience of reference, these data have been tabulated in table 3. It is to be remembered that these figures are based on only 704 specimens and additional material may modify them somewhat; but it is probable that they are adequate to establish the essential character of these relations.

Foot Length

Table 4. Foot length and its proportion to sitting height.

For purposes of determining age, neither the foot length nor the head size, when used alone, has as much value as do weight and sitting height. The head size shows considerable variation, due in part to the mechanical molding which occurs in the great majority of specimens and in part to the varying effect of formalin on the soft tissues of the scalp, the distention of which affects the size reading considerably more than the corresponding increase in the sitting height. The foot length, as compared with the sitting height, possesses the disadvantage of being smaller and having a smaller weekly increment. Nevertheless, these two measurements serve as additional controls, and in cases of dismembered specimens they often constitute the only reliable criteria for the determination of the age. From an uninjured foot one can determine fairly closely the normal sitting height, weight, and menstrual age of the specimen by means of the respective correlation curves.

Accurate measurement of the foot can not be made earlier than in embryos about 24 mm. long. Briefly stated, the mean foot length at different intervals is as follows: at 8^ weeks, about 4 mm.; end of the eleventh week, about 7 mm.; end of the fourteenth week, 14 mm. ; end of the sixteenth week, about 20 mm. ; end of the twenty-second week, about 40 mm.; end of the thirtieth week, about 60 mm.; and at term it averages 82.5 mm. If the foot length is plotted as a curve at weekly intervals from the eighth week to term it will be found that the growth is relatively slow at first and does not reach its maximum weekly increment until about the fourteenth week, after which there is a weekly increase of about 3 mm., continuing with slight variation to term. The growth is a little more rapid from the fourteenth week (CR 87 mm.) to the twenty-sixth week (CR 228 mm.) and a little less rapid from then until term.

The foot length of our individual specimens has been entered in the form of dots on charts 1, 3, and 5, showing the length in millimeters correlated to both sitting height and weight. The pathway occupied by the dots, aside from a few outlying ones, is smooth and distinct. When the upper and lower margins of this pathway are outlined it gives the maximum and minumum foot lengths at the different intervals. These readings are entered in the accompanying table 4. The average of the maximum and minumum lengths is entered as the mean foot length. There is also entered in the same table the percentage of the sitting height formed by the mean foot length for the end of each week. Examination of this table shows that there is a gradual increase in the length of the foot relative to the length of the embryo. This does not begin, however, until the fetus is about 70 mm. long. From 30 mm. to 60 mm. the foot grows less rapidly than the body as a whole. Beyond 70 mm. there is a slight gradual increase in the length of the foot in proportion to sitting height, so that during the last month the foot length is 23 per cent of the sitting height, having increased from about 15 per cent, which existed in fetuses 70 mm. long. This increase in the percentage of the foot length to sitting height is not so much due to an acceleration in the growth of the foot as to a retardation in the increase in sitting height that characterizes the latter part of fetal growth.

Head Modulus

As an index of head size a modulus was selected, consisting of the mean of the greatest horizontal circumference of the head and the biauricular transverse arc (i. e., the distance from one external auditory meatus over the vertex of the head to the other external auditory meatus). The sum of these, divided by 2, is taken as a head modulus and expressed in millimeters. These measurements were selected because they can be systematically and accurately made for specimens from 50 mm. long to term, and because they tend to correct each other when the head is molded, as is frequently the case. If I were going to make these measurements again I should add the biauricular diameter to the transverse arc, thus obtaining a transverse circumference of the head. The average between this and the greatest horizontal circumference would approximate the average circumference of the head. Dr. Schultz, of this laboratory, in his anthropological study of the fetal period, is utilizing the horizontal, transverse, and sagittal circumferences, the latter two being composed respectively of the transverse arc and biauricular diameter, and the sagittal arc and nasion-inion diameter. By dividing his results by 3 he obtains a still more accurate mean head circumference. It is probable, however, that the head modulus as used may be relied upon as giving an approximate index of the normal head size and the essential proportions of the latter to body length.

Table 5. Head modulus, representing head size in millimeters, based on the mean between greatest horizontal circumference and the biauricular transverse arc.

The head-modulus data for the individual specimens were entered as dots on charts 3 and 5, showing their correlation to the mean sitting height and weight. The dots do not form as compact a cluster as those for sitting height in the corresponding charts 2 and 4; there is, however, a consistent pathway through which a smooth curve could be drawn, showing the mean head modulus. The readings taken from this curve are entered in table 5. Up to the thirty-second week the plotted field could be definitely outlined and its upper and lower limits for the successive weeks were entered in the same table, the few widely outlying entries beingomitted. Beyond the thirty-second week there were enough data to complete a mean curve up to term, but not enough to warrant a delimitation of the range of variation. In table 5 were also entered the weekly increments in the mean head modulus and the proportion of the head modulus to the sitting height, expressed in percentages.

From an examination of the table, it will be seen that the actual increment in head size undergoes a slight gradual decline from the twelfth to the fortieth week, decreasing from 13 mm. to 5 mm. Expressed in terms of percentage the decrease is greater, being from 20 per cent to about 2 per cent. As in the case with sitting height, the greatest increments are before the eighteenth week. The decrease in the percentage increment is a little greater in the head modulus than in the sitting height, and its mean curve, as seen in charts 3 and 5, gradually recedes from the mean curve of the latter. On this account the head modulus, though at first nearly equivalent to the sitting height (96 per cent), toward the end of pregnancy is less than 84 per cent. In a few outlying instances the head modulus equals or exceeds the sitting height. These are cases showing unusual distention of the soft tissues of the seal]). Under the usual conditions, if the head modulus is greater than the sitting height it should be regarded as abnormal. In this way we may be able to detect early cases of hydrocephalus. On the other hand, if the head modulus is too small we have an indication of microcephalus.


Mall FP. VIII. Determination of the age of human embryos and fetuses in Keibel F. and Mall FP. Manual of Human Embryology I. (1910) J. B. Lippincott Company, Philadelphia.

Mall FP. On the age of human embryos. (1918) Amer. J Anat. 23: 397-422.

Meyer AW. Curves of prenatal growth and autocatalysis. (1914) Arch. f. Entwicklungsmechanik der Organismen, vol. 40, pp. 497-525.

— . 1915. Fields, graphs, and other data on fetal growth. Contributions to Embryology, vol. 2, pp. 55-68, Carnegie Inst. Wash. Pub. 222.

Riggs, T. F., 1904. A comparative study of white and negro pelves, with a consideration of the size of the child and its relation to presentation and character of labor in the two races. Johns Hopkins Hosp. Reports, vol. 12, pp. 421-454.

Schultz, A. H, 1919. Changes in fetuses due to formalin preservation. Amer. Jour. Phys. Anthropol., vol. 2, pp. 35-41.

— , 1920. An apparatus for measuring the newborn. Johns Hopkins Hosp. Bull., vol. 31.

Zangemeister, W., 1911. Die Altersbestimmung des Foetus nach graphischer Methode. Zeitschr. f. Geb. u. Gyn., vol. 69, pp. 127-142.

Table 6 and 7

Table 6. Data on embryos weighing less than 10 grams.

Tables 6 and 7 constitute a list of normal embryos of the Carnegie Collection on which the correlation curves of sitting height, weight and menstrual age are based.

The material is arranged according to weight, i.e., its formalin weight after being in a 10 per cent solution about two weeks.

Some of the specimens of over 400 grams in weight were injected with the formalin solution through the umbilical vessels, as noted under "Remarks"; in such cases the weight given is the fresh weight plus 5 per cent, this being more nearly comparable to the weight of specimens simply immersed in the solution.

The sitting height of specimens 50 mm. long and over was measured with the body held as straight as possible. Specimens smaller than this were usually measured in their natural curved posture. In the overlapping field both measurements are given, as seen in table 6.

By head modulus is to be understood the mean between the greatest horizontal circumference of the head and the biauricular transverse arc .

The estimated menstrual ages of the specimens are based on the curve in chart 6, with the exception of the first 8 specimens in table 6, the ages of which are based on the estimates of Mall (1910).

Table 7. Data on embryos weighing 10 grams or more.


Chart 1

Chart 1. Growth curve based on crown-rump length and weight of embryos weighing less than 20 grams. Each specimen measured with the body straightened out is entered as a dot; those measured in their natural curved posture are entered as crosses.

The mean curves drawn through these two fields overlap, showing that if embryos of 36 to 42 mm. are straightened out it adds about 4 or 5 mm. to their length.

Toward the bottom of the chart the foot length in millimeters is plotted according to weight of embryo.

A scale showing menstrual age, as in the succeeding charts, is placed at the top of the chart.

This scale is based on chart 6.

Chart 1. Growth curve based on crown-rump length and weight of embryos weighing less than 20 grams.

Chart 2

Chart 2. Mean curve of growth of the Carnegie Collection weighing less than 400 grams, based on sitting height and weight. Chart 2. Mean curve of growth of the Carnegie Collection weighing less than 400 grams, based on sitting height and weight.

Chart 3

Chart 3. Foot length and head modulus correlated to sitting height and weight for fetuses under 400 grams in weight.

The heavy line represents the curve of the mean head modulus; the broken line is the mean sitting height or crown-rump length, taken from chart 2.

Chart 3. Foot length and head modulus correlated to sitting height and weight for fetuses under 400 grams in weight.

Chart 4

Chart 4. Carnegie Collection over 200 grams in weight, plotted for silting height and weight. Chart 4.  Carnegie Collection over 200 grams in weight, plotted for silting height and weight.

Chart 5

Chart 5. Field and curve of mean menstrual age for specimens in the Carnegie Collection having menstrual histories.

The circles placed at the 32d, 36th, and 40th weeks are taken from Zangemeister's (1911) data and are used to complete the curve to the termination of pregnancy.

Chart 5. Field and curve of mean menstrual age for specimens in the Carnegie Collection having menstrual histories

Glossary Links

Glossary: A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z | Numbers | Symbols | Term Link

Cite this page: Hill, M.A. (2018, December 11) Embryology Book - Contributions to Embryology Carnegie Institution No.55. Retrieved from

What Links Here?
© Dr Mark Hill 2018, UNSW Embryology ISBN: 978 0 7334 2609 4 - UNSW CRICOS Provider Code No. 00098G