International Endedmticjminuil (1990) 23,34-39
Root canal anatomy of mandibular second molars. Part I
S. A. MANNl'NG Department ofRestorative Dentistry, Faculty of Dental Science,
University of Melbourne, Melboume, Victoria, Australia
Summary. The root canal anatomy of 149 man-
dibular second molars was studied using a technique
in which the pulp was removed, Ehe cana! space filled
with black ink and the roots demineraiized and made
transparent. Ofthe 149 teeth, 22 percent had single
roots, 76 per cent had two roots and 2 per cent had
three roots. In the single-rooted teeth, three canals
were most common, while in the mesial root ofthe
two-rooted teeth, tn'o canals that joined near the
apex and one canai in the distal root were most
frequent. Round canals were most common in two-
rooted teeth and C-shaped canals in single-rooted
teeth. Transverse anastomoses were found in 33 per
cent of roots, most conamoniy in the middle third of
the root. Lateral canals were found in 72 per cent of
roots, most commonly in the apical third ofthe root.
The apical foramen was positioned at the apex in
only 33 per cent of roots, and apical deltas were
found in 35 per cent.
The patient's age and race aflfected canal shape,
with more round canals present in patients over 35
years of age, and more C-shaped canals in Asians.
The sex of the patient and the side of the mouth
affected the presence of apical deltas, with more
being found in males and on the left side. Single-
rooted teeth had more complex root canal systems
than two-rooted teeth, with more lateral canals,
transverse anastomoses, apical deltas and C-shaped
canals.
Introduction
A general trend towards the retention of teeth
rather than extraction is evident today, and
consequently mandibutar second molars are
now tnore frequently involved in root canal
treatment (Seltzer 1987). A thorough ktiowl-
edge ofthe root canal anatomy ofteeth involved
in such treatment is required for successful
debridement and obturation ofthe root canal
system.
The anatomy of root canals is complex and
difficult to study. In order to view root canals.
Correspondence: Mr S. A. Manning, 1 Lister
House, 11-12 Wimpole Street, London WIM 7AB,
UK.
34
the calcified root tissue must either be removed
or be rendered transparent in some way.
Different methods have been used by various
investigators to overcome this problem, and
have been well summarized by Manning (1987)
and Walker (1988a). The latter concluded that
studies in which radiographic and clearing
techniques were used had considerable merit.
While radiography may demonstrate the main
canals within the root, it is unlikely to show the
complexities of root canal anatomy. Clearing
techniques which render the teeth transparent
proNide the most reliable method for viewing
the entire root canal system, and such tech-
niques have enabled identification of lateral
canais, transverse anastomoses, apical deltas
and other canal complexities (Okumura 1927,
Fitz-Walter 1981, Vertucci 1984). Variations
of this method include different initial tooth
storage media, techniques for removing the
pulp, and materials used to highlight the canal
system (Fitz-Walter 1981). The techniques
used in this study were initially developed at
the University of Melbourne (Fitz-Walter
1981) and were slightly modified (Manning
1987).
The root canal anatomy of mandibular
second molars has been described by a number
of investigators (Hess 1925, Pederson 1949,
Tratman 1950, Ainamo & Loe 1968, Pineda &
Kuttler 1972, Green 1973, Tamse & KaflFe
1981, Vertucci 1984, Kotoku 1985, Yingetal.
1988, Walker 1988b, Weine et al. 1988).
Conflicting results have been obtained, which
may be due to differing methods of study, or
to variation in the population from which the
teeth were taken with regard to racial group,
age, sex and the side ofthe mouth from which
the tooth originated. The main areas ofdispute
are the number ofroots possnsed by mandibu-
lar second molare, and the type Mid shape
of canals. A higher number of mandibular
second molars with single roots has been

found in Mongoloid populations (Kotoku
1985, Walker 1988b); and C-shaped canals
have also been found more frequently in the
same populations (Kotoku 1985, Yang et al.
1988, Walker 1988b).
Canal types have been described as type 1,
2, 3 or 4 (Vertucci 1984), where:
(i) a type I canal system has a single canal
from orifice to apex;
(ii) a type 2 canal system has two canals that
combine and exit as one at the apex;
(iii) a type 3 canal system has two separate
canals from orifice to apex and;
(iv) a type 4 system has one canal that separ-
ates and exits as two at the apex.
This definition of canal type does not accom-
modate more than two canals in a root, or any
combination other than two canais into one, or
one into two. Instead, a simple definition is
used in this study, and explained in the
methods section.
The aim of this study was to investigate the
root canal system of mandibular second molars
and its relationship to the patient's age, race,
sex and side ofthe mouth.
Materials and methods
One hundred and forty-nine mandihular
second molars were collected and stored in
water immediately after extraction. Infor-
mation concerning the patient's age, sex, race
(Caucasian, Asian, Indian or other), and the
side of the mouth from which the tooth
originated, was recorded at the same time.
After 4 weeks of storage in water, the teeth
were decoronated at the amelo-cemental junc-
tion and stored for another 1-A weeks in water,
so that autolysis and maceration of the pulp
could continue. The roots were then soaked in
a 2 per cent ficin solution' at 37°C for 14 days
to remove pulpal tissue. Ficin is derived from
fig tree latex and its proteolytic action assists in
the removal of organic tissue. The teeth were
washed thoroughly, and black ink^ with a
10 per cent black gelatine solution was then
introduced into the canal system and fixed in
place with formalin. Foilowing this, the roots
were demineralized in a decalcifying solution
' Calhiochem, Los Angeles, CA, USA.
'Wiltshire's Penguin Indian Ink, Melbourne,
Australia.
Root canals ofmandibutar second molars 35
(4mol/l formate buffer, pH 2.3) for 7 days.
They were then dehydrated in ascending
griules of alcohol (70 per cent, 90 per cent,
](K) per cent).
The roots were cleared in cedar wood oil^
for 1-2 days, prior to examination with a stereo
dissecting microscope (magnification x 20).
Data were collected according to the following
categories: number of roots; canal type; apical
foramen position; canai shape; presence and
position of transverse anastomoses; presence,
number and position of lateral canals; presence
ofa delta ofcanals at the apex. The canal system
was defined as follows: type 1, in which a single
canal is present; type 2-1, in which there are
two canals coronally that combine and exit as
one at the apex; type 2, in which there are two
separate canals from orifice to apex; type 1—2
in which there is one canal that separates and
exits as two canals at the apex, and so on. A
t\'pe 3 canal system describes three canals in a
root that may or may not combine before they
exit at the apex. Chi-squared tests were used
to investigate any relationship between the
variables.
Results
The patient's age at the time of tooth extrac-
tion was in the range 15-88 years. The mean
age was 43.1 years (SD 18.2). Race was
recorded for 123 patients, but not for the
remaining 26 patients. Ninety-nine teeth were
from Caucasian patients, 14 teeth were from
Asian patients, and 10 teeth were from other
racial groups. No teeth were collected from
Indian patients. Ninety-one teeth were col-
lected from male patients and 57 teeth from
female patients. The sex of one patient was not
recorded. Sixty-three teeth were from the left
side and 86 teeth were from the right side of
the mouth.
Of the 149 mandibular second molars
studied, 33 teeth had one root, 113 teeth had
two roots and three teeth had three roots. AU
the three-rooted teeth had a bifurcated mesial
root and one distal root, and were thus classed
as two-rooted in order to simplify analysis of
the results. Therefore a total of 265 roots were
studied, 33 from 33 single-rooted teeth and
232 from 116 two-rooted teeth. Of the 265
'BushBoake Allen, Melbourne, Australia.
J6 S.A. Manmag
Table I. Number of roots with round, oval or C-
shaped canals
Round Oval C-shaped Total
Single root 10 4 19 33
Mesial root 59 57 0 116
Distal root 87 29 0 116
Total 156 90 19 265
roots examined, 156 had round canals, 90 had
oval canals and 19 had G-shaped canals. The
single-rooted teeth had 10 round canals, four
oval canals and 19 C-shaped canals. The
mesial root of two-rooted teeth had 59 round
canals attd 57 oval canals, while the distal canal
had 87 round canals and 29 oval canals. The
C-shaped canals were not found in two-rooted
teeth (Table I).
In single-rooted teeth, catia) type 3 oc-
curred most frequently, followed by canal
types 2-1, 2, 1, 1-2-1 and 1-2. In the mesial
root of two-rooted teeth, canal type 2-1 oc-
curred most frequently, followed by types 1,
1-2-1, 2, 1-2 and J. In the distal root, canal
type 1 (a single canal) was present in nearly ail
specimens. Types 1-2-1 and 1-2 were found
infrequently (Table II).
The apical foramen was positioned at the
apex infiveout of the 33 single-rooted teeth, in
44 out of 116 mesial roots, and in 31 out of 116
distal roots. Of the 265 roots examined, 80
showed the apical foramen to be at the apex.
Five hundred and thirty-one lateral canals
were found in 191 roots. In 74 roots, lateral
canals were not found. However, they were
found in 28 of the 33 single-rooted specimens,
96 of the 116 mesial roots and in 67 of the 116
distal roots. Lateral canals were positioned
most frequently in the apical third of the root
(59.4 per cent), followed by the furcation area
(22.6 per cent), then the middle (16.7 per cent)
and coronal third (1.3 per cent) of the root.
Transverse anastomoses were found in 87 out
of 265 roots. They occurred in 21 of the 33
single-rooted teeth, 63 of the 116 mesial roots,
and three of the 116 distal roots. They were
most frequently found in the tniddie and then
in the apical third of the roots. Apical deltas
were seen in 24 single roots, 40 mesial roots
and 30 distal roots, with 94 of the 265 roots
examined having apical deltas.
Statistical analysis
Atialysis of the results using Chi-squared tests
yielded the following information:
(i) canal shape was significantly related
to age (P<0.01), with more oval canals
observed in individuals under 35 years
of age and mm« round canals in those
over 35 years of age;
(ii) canal shape was also significantly related
to race (P<0.02), with more C-shaped
canals observed in teeth from Asians;
(iii) the presence of apical deltas was signifi-
cantly related to the sex of the individual
(P<0.01), with more teeth from males
than females having apical deltas;
(iv) the presence of apical deltas was signifi-
cantly related to the side of the
mouth from which the tooth originated
(/'<0.01), with more in the teeth from
the left side of the mouth;
(v) mandibuiar second molars with single
roots generally had a more com-
plex root canal system than two-
rooted specimens, more lateral canals
(P<O.Ol) and transverse anastomoses
(P < 0.01) were observed, and the apical
foramen was found at the apex less
frequently (/>< 0.05).
Discussion
Previously reported differences in the root
canal anatomy of mandibular second molars
may be due to the population from which the
teeth were chosen. If a Caucasian population
were studied, then predominantly two-rooted
mandibutar second molars could be expected
(Ainamo & Loe 1968, Tamse & Kaffe 1981,
Vertucci 1984, Weine et al. 1988), whereas
if an Asian or Mongoloid population were
examined then a higher number of teeth would
be single-rooted (Kotoku 1985, Walker
1988b). In this study, single-rooted mandibu-
lar molars were found in both Caucasians and
Asians; the latter showed a higher frequency,
but this was not significant.
In two-rooted mandibular second molars,
the canal types most frequently found in
previous studies were type 1 in the distai root
and types 2-1, 1, 2, 1-2-1 and 1-2 in the
 Root canals afmandihilar second matars 37
Table II. Number of roots with different types of canal

Canal type*

1 2-1 1-2-1 2 1-2 3

Single root 3 6 3 5 3 13
Mesial root 26 37 21 17 11 3
Distal root 113 1 2

•Canal type: 1, single canal; 2-1, two canals that join apicalty; 1-2-1, one
canal that divides but rejoins apicalty; 2, two separate canals; 1-2, one canal that
divides apically; 3, three separate canals.

Table HI. Comparison of canal types found in two-rooted mandibuiar second molars

Percent;^e of canal types

Author Root 1 2-1 1-2-1 2 1-2 3

Pineda &Kuttler( 1972) Mesial 58 20.6 13.8 7.6

Distal 94 2.5 — 3.0 0.5
Green (1973) Mesial 13 49 — 38 — —
Distal 92 5 — 3 —
Vertucci (1984) Mesial 27 38 — 26 9 —
Distal 42 3 — 4 1 —
Manning (1987) Mesial 22.5 32.7 IS.! H.6 9.5 2.6
Distal 97.4 0.9 1.7
Weine rt a/. (1988) Mesial 4.0 52.0 — 40.0 —
Distal 55.3 9.3 — 1.3 — —

mesial root (Table III). In general, two-rooted
mandibular second molars have a single distal
canal (more than 90 per cent) and two mesial
canals that may combine (more than 70 per
cent). A radit^raphic study of two-rooted
mandibular second molars from a Mexican
population demonstrated a much higher per-
centage (58 per cent) of type 1 canals in the
mesial root (Pineda & Kuttler, 1972). This
may have been due to the teeth being of
Mongoloid origin, as has been suggested by
Walker (1988b), but was not noted in this
study.
Single-rooted mandibular second molars
have been reported less frequently (Kotoku
1985, Yang et al. 1988, Walker 1988b, Weine
et al. 1988), and the canal types have not
been fully investigated. A radiographic study
(Weine et al. 1988) of three single-rooted
specimens found one to have a single canal and
two to be C-shaped, and photographs from
another study (Walker 1988b) of selected
specimens that were cleared showed the com-
plex anatomy characteristic of single-rooted
specimens. In the current investigation a
variety of canal types was noted, with type 3
being the most frequent (Table II).
Canal shape in mandibular second molars
has not been fully investigated previously.
The finding or incidence of C-shaped canak
has only been mentioned (Cooke & Cox, 1979,
Kotoku 1985, Walker 1988b, Weine et al.
1988, Yang et at. 1988). In this study the canal
shape was significantly related to race, with
more Asians having G-shaped canals. The
low number of Asians in the sample may not
38 5. A. Matamf
allow valid comparisons to be made with the
population as a whole, but C-shaped root
canals have been noted regularly in Hong
Kong Chinese (Walker 1988b), Chinese (Ti'ang
et al. 1988), Japanese (Kotoku 1985), East
Greenland Eskimo (Pederson 1949) and
Malaysian Asian (Tratman 1950) populations.
These observations would appear to support
the findings of the present study. In studies
of non-Asian or non-Mongoloid popuiatiotis,
C-shaped root canals in mandibular second
molars have rarely been mentioned. The
C-shaped canals are discussed fully in Part II
of this paper (Manning 1990).
Lateral canals, transverse anastomoses and
apical deltas all contribute to the complexities
of the root canal system. Lateral canals were
found in a higher proportion of mandibular
second molars in this study than in previous
investigations (Pineda & Kuttler 1972,
Vertucci 1984, Borio er al. 1985); they were
found most frequently in the apical third of the
root. Transverse anastomoses were found less
frequently than lateral canals, and were most
common in the middle third of the root. Apical
deltas have been noted in previous studies
(Pineda & Kuttler 1972, Vertucci 1984).
The effect of age on the root canal system
has been studied by a number of investigators
(Hess 1925, Barrett 1925, Philippas 1961,
Pineda & Kuttler 1972, Fitz-Walter 1981),
and is recognized by most clinicians. Large
ovoid or dumb-bell shaped canals tend increas-
ingly to resemble two round canais with age
and further deposition of dentine, especially in
the mesial roots of lower molars, and in lower
incisors. This was confirmed in the current
study, where it was found that there were more
oval canals in teeth from individuals under 35
years of age and more round canals in individ-
uals over 35 years. The deposition of dentine
with time resulted in oval canals developing
into two round canals in mesial roots, or one
round canal in distal roots. There was no
relationship between age and the incidence of
lateral canals and apical deltas, as has been
reported by some investigators (Barrett 1925,
Hess 1925).
Sexual dimorphism ofteeth has been studied
by a number of investigators (Morrees 1957,
Gam et al. 1965, Rosenweig 1970, Ditch &
Rose 1972). Root length was found to be related
to sexual dimorphism (Gam et al. 1965), with
male txeth tending to be larger and wider, with
longer roots. Females showed a higher inci-
dence of single-rooted tnandibular second
molars than males in a sample of teeth from a
Japanese population (Kotoku 1985). However,
sexual dimorphism was low or absent in studies
of the native American populations (Turner
1984). In the current study, there was a signifi-
cantly higher incidence of apical deltas in male
teeth compared with teeth from females. The
incidence of apical deltas in relation to the sex
of the individual has not been investigated
previously.
The relationship of the side of the mouth
from which the teeth originated with root
canal anatomy has also not been investigated
before. A higher incidence of apical deltas on
the left side of the mouth was noted in this
study.
Lateral canals, transverse anastomoses and
apical deltas were most frequently found in
single-rooted specimens, then in the mesial
root, and least frequently of ail in the distal
root of two-rooted specimens. In addition, all
C-shaped canals were found in single-rooted
teeth. It can be concluded that single-rooted
mandibular second molars have a more com-
plex root canal system than two-rooted teeth,
and that mesial roots have a more complex
system than distal roots.
References
AINAMO, J. & LOE, H . (1968) A stereomicroscopic
investigation of the anatomy of the root apices of
910 maxillary and mandibular teeth. Odontolo-
giska Foreningens Tidskrift, 76,417-426.
BASRETT, M . T . (1925) The internal anatomy of the
teeth with special reference to the pulp and its
branches. Dental Cosmos, 67,581-592.
BORIO, P.S., BERHONE, S. MINA, M . , TARELLO, F .
GIORDANO, M . & BARBERO, P. (1985) Studio
sulla presenza di forami accessori nei molari
permanenti. Considerazioni su 100 casi. Minerva
Stomatoiogica, 34,347-350.
CooKE, H.G. & Cox, F.L. (1979) C-shaped canal
conjurations in mandibular molais. Journal of
the American Dental Association, 99,836-839.
DITCH, L . E . & R o a , J.C. (1972) A multivariate
dental sexing technique. Jtmmai of Physical
Anthropolegy, 37,61-64.
FITZ-WALTER, P.B. (1981) The root outal mor-
phology of mandibular premolars. MDSc Thesis,
University of Melbourne, Australia.

GAHN, S . M . , LEWIS, A.B. & KERESKY, R.S. (1965)
X-linlced inheritance of tooth size. Journal of
Dental Research, 44,43^-441.
GREEN, S. (1973) Double canals in single roots.
Oral Surgery, Oral Medicine and Oral Pathology,
35,689-696.
HESS, W. (1925) The Anatomy ofthe Root Canals of
the Teeth of Permanent Dentition. Part I. John
Bale and Sons, London.
KOTOKU, K . (1985) Morphological studies on the
roots ofthe Japanese mandibular second molars.
Shikwa Gakuho, 85,43-64.
MANNING, S..^. (1987) The root canal morphology
of mandibular second molars MDSc Thesis,
University of Melboume, .\ustralia.
.MANNING, S..^. (1990) Root canal anatomy of
mandibutar second molars. Part I!. C-shaped
canals. Intemationat Endodontic Journal, 23,
40-45.
MORREES, C.F.A. (1957) The Ateut dentition: a
correlative study of dental characteristics in an
Eskimoidpeople, pp. 77-101. Harvard University
Press, Cambridge, Mass.
OKUMURA, T . (1927) Anatomy ofthe root canals.
Joumat of the American Dental Association, 14,
632-636.
PEDERSON, P.O. (1949) The East Greenland Eskimo
Dentition: Numerical Variations and Anatomy.
Bianco Lunos Bogtrykkeri, Copenhagen.
PHILIPPAS, G.G. (1961) Influence of occlusai wear
and age on formation of dentine and size of
pulp chamber. Journal of Dental Research, 40,
1186-1198.
PINEDA, F . & KUTTLER, Y. (1972) Mesiodistal and
buccoiingual roentgenographic investigation of
7275 root canals. Oral Surgery, Oral Medicine
and Oral Pathology 33,101-110.
Root canals of mandibular second molars 39
G, K.A. (1970) Tooth form as a dis-
tinguishing trait between sexes and human
populations. Joumat of Dental Research, 49,
1423-1426.
SELTZER, S . (1987) Introduction. In Pathways of
the Pulp, (eds S. Cohen and R.C. Bums), 4th
edn, pp. !-2. CV. Mosby Co. St. Louis.
TAMSE, K & KAFFE, I. (1981) Radiographic survey
ofthe prevalance of conical lower second molar.
Intemational Endodontic Journal, 14,188-190.
TRATMAN, E . K . (1950) A comparison of teeth of
people; Indo-European racial stock with the
Mongoloid racial stock. Dental Record, 70,
63-88.
TURNER, C . G . (1984) Advances in the dental
search for Native American origins. Ada
Anthropogenetica, 8,23-78.
VERTUCCI, F.J. (1984) Root cana! anatomy ofthe
human permanent teeth. Oral Surgery, Oral
Medtcme and Oral Pathatogy, 58, 589-599.
WALKER, R.T. (1988a) The root canal anatomy
of mandibular incisors in a Southern Chinese
population. International Endodontic Joumat, 21,
218-223.
WALKER, R . T . (1988b) Root form and canal anat-
omy of mandibular second molars in a Southem
Chinese popuiaiicirL Joumat of Endodontics, 14,
325-329.
WEINE, F.S., PASIEWICZ, R.A. & Rici, R.T. (1988)
Canal configuration of the mandibular second
molar using a clinically oriented in vitro method.
Joumat of Endodontics, 14,207-213.
YANO, Z.P., YA.NG, S.F., LIN, Y.C., SHAY, J.C. &
CHI, C.Y. (1988) C-shaped root canals in
mandibular second molai^ in a Chinese popu-
lation. Endodontics and Dentat Traumatotog)', 4,
160-163.