Biology Reference
In-Depth Information
vertebrae (T5
e
7) and the first through third lumbar vertebrae (L1
e
3) best correlate with the
overall spine length (r
0.952). They concluded that this could be used if the entire vertebral
column is not available.
Fully and Pineau (1960)
then compared different combinations of
long bone and trunk lengths, developing equations for the femur and five lumbar vertebrae
(r
¼
0.908) (
Fully and Pineau, 1960
).
Fully and Pineau measured the long bones differently from Trotter and Gleser, so compara-
tive studies between the two are not possible. None of these regressions, except the combined
vertebral lengths as per
Fully and Pineau (1960)
, is as powerful as the formulae from the long
bones. Tibbetts (1981) articulated the vertebral column to estimate stature, which is recom-
mended only if the long bones are not available. However, any bones that do not represent
a proportion of an individual's living stature should be avoided (
Trotter and Gleser, 1958
;
Rathbun, 1984; Joyce and Stover, 1991).
¼
0.926) and for the tibia and five lumbar vertebrae (r
¼
Stature Regression for Non-U.S. Populations
Stature regression formulae have been calculated for populations all over the world.
A number of studies have estimated the stature of Europeans. One early study measured
the stature of 3000 living criminals in England from all over Great Britain (
Macdonnel,
1901
). This study measured head height, head breadth, third finger length, cubit length
(elbow to the tip of the middle finger), and foot length.
Brietinger (1937)
measured the
maximum lengths of long bones from 2400 German male skeletons, which estimated a taller
stature than did Pearson's or Manouvrier's equations.
Mendes-Correa (1932)
measured the
maximum lengths of the humerus, ulna, radius, femur, tibia, and fibula from a Portuguese
sample. By studying the stature of both cadavers and living individuals, Mendes-Correa
determined that living stature is 2 cm shorter than that of cadavers.
In 1950, Telkk¨ measured a Finnish sample of skeletons, which included 115 males and 39
females (
Te lkk¨,1950
). Telkk¨ used the maximum length of the humerus, femur, tibia, and fibula
and the physiologic lengths of the radius and ulna. He reported that supine cadaver stature is 2
cm greater than living stature. Furthermore, he recommended that if the bones are measured
when wet, then 2 mm must be subtracted from the bone length before using the formulae due
to shrinkage in drying (
Te lkk¨, 1950
).
Allbrook (1961)
measured the percutaneous lengths (i.e.
via palpable landmarks on the skin) of the tibia and ulna from fleshed individuals with the limbs
semi-flexed of British and East African ancestry.
Cern
´
and Komenda (1982)
measured the
humerus and femur of a Czech sample that included 148 males and 104 females.
Correctly or incorrectly, a common practice in stature estimation has been to develop
regression formulae for populations from Asia lumped together with Native American pop-
ulations. Stewart (1954) used the data from
Trotter and Gleser (1952
) and developed his own
regression equations for Asians. Neumann (1967) prepared formulae for males and females
from the femur and tibia for two Native American populations, but provided no standard
error estimate. Not content with the accuracy of Trotter and Gleser formulae for Mesoamer-
icans,
Genov´s (1967)
produced stature regression formulae from 76 males and 59 females of
Indigenous Mexican,
Mestizo
, and European ancestry using all of the long bones.
7
Genov´s
7
Try to pay attention to the size of the sample in each study. In the study by
Genov
´
s (1967)
, the number of
males (76) and females (59) was further broken down into three ancestry groups, which is statistically
problematic.
