INTRODUCTION
Medicinal plants traditionally occupied an important
position in rural and tribal lives of India and are considered as
one of the most important sources of medicines since the
dawn of human civilization. One such an important medicinal
plant is medicinal coleus (Coleus forskohlii Briq.). The
tuberous roots of coleus are rich source of forskolin, a
diterpenoid activates Cyclic Adenosine Monophosphate or
AMP in the cells (Bhat et al, 1977). Coleus forskohlii is the
only known natural source of forskolin. Due to its multifaceted
pharmacological effects, forskolin is used for treatment of
eczema (atopic dermatitis), asthma, psoriasis, cardiovascular
disorders and hypertension, where decreased intracellular
cAMP level is believed to be a major factor in the development
of disease process (Rupp et al, 1986). Extent of genetic
variation in Coleus forskohlii is limited. Continuous
vegetative propagation for many years has reduced the vigour
and tolerance to biotic and abiotic stress, causing low yields.
Hence, yield and quality is enhanced possibly by developing
Studies on correlation and path analysis in mutants of Coleus (Coleus forskohlii Briq.)
for yield and forskolin content in V
2
M
1
generation
M. Velmurugan, K. Rajamani, P. Paramaguru, R.Gnanam1 and J.R. Kannan Bapu2
Department of Spices, Plantation crops and Medicinal plant Unit
Tamil Nadu Agricultural University, Coimbatore – 641 003, India
E-mail: hortmrvelu@yahoo.com
ABSTRACT
The present investigation was carried out during 2003-2007 involving terminal cuttings of coleus ecotype ‘Garmai’.
Genotypic correlation coefficient between yield and its components in mutants of coleus was studied, viz., plant height,
number of branches plant-1, number of leaves plant-1, number of tubers plant-1, tuber length and tuber girth were
found to have positive and highly significant correlation with yield. However, forskolin and essential oil content
showed negative correlation with yield. Path analysis of component characters on yield of Coleus in V
2
M
1
generation
exerted positive direct effect through the characters plant height, number of leaves plant-1 and number of tubers plant-
1. Similarly, direct effect was observed to be negative through number of branches plant-1 (-0.930), total amount of
alkaloids (-0.066) and forskolin content (-0.026). The current investigation resulted in residual effect of 0.158
indicating the accuracy and appropriate selection of component character for crop improvement programme. Weightage
must be given to component characters exhibiting positive attributes towards fresh tuber yield in Coleus. However,
some traits with negative attributes are also chosen for getting improved quality, i.e., forskolin content, without much
inhibition on fresh tuber yield plant-1.
Key words: Coleus forskohlii, correlation, Path analysis
a mutant in this species with high tuber yield and improved
forskolin content through induced mutations. The ultimate goal
of crop improvement in coleus is to improved tuber yield and
forskolin content. Being a complex trait, the tuber yield is
largely influenced by many component characters. Information
on strength and direction of correlation of these component
characters on tuber yield and inter se association among them
would be useful in designing breeding programmes for yield
improvement. The relationship between yield and its
component characters is likely to vary according to the genetic
material used and environment under which the material is
evaluated as well as due to interaction of these factors.
Therefore, it is worthwhile to study the heritable association
between variables (Genotypic correlation) for identification
of important yield components so that weightage can be given
to these characters of importance in further breeding
programmes (Johnson et al, 1955). The current investigation
confines to correlation and path analysis in mutants of coleus
in V
2
M
1
generation.
Present address: 1 Agricultural College and Research Institute, Madurai
2 Agricultural Research Station, Aliyarnagar
J. Hortl. Sci.
Vol. 4 (1): 63-67, 2009
64
MATERIAL AND METHODS
The present investigation was carried out at
Medicinal plant unit, Horticultural College and Research
Institute, Tamil Nadu Agricultural University, Coimbatore
during 2003-2007. Terminal cuttings of coleus ecotype
‘Garmai’ was obtained from Manjini in Salem district of
Tamil Nadu, where this crop is grown commercially by the
farmers in a larger extent. Based on preliminary
experiments, it is concluded that the LD
50
value for gamma
rays was 3.00 kR and for EMS it was noticed at 1.00 %
concentration which was exposed for a period of 3.00 h.
Based on the sensitivity studies, mutagenic treatments were
formulated viz., AT
1
– Control, AT
2
- 2.50 kR gamma rays,
AT
3
- 3.00 kR gamma rays, AT
4
- 3.50 kR gamma rays, AT
5
- 0.50 % EMS, AT
6
- 1.00 % EMS, AT
7
- 1.50 % EMS, AT
8
- 2.50 kR gamma rays + 0.50% EMS, AT
9
- 2.50 kR gamma
rays + 1.00% EMS, AT
10
- 2.50 kR gamma rays + 1.50%
EMS, AT
11
- 3.00 kR gamma rays + 0.50% EMS, AT
12
-
3.00 kR gamma rays + 1.00% EMS, AT
13
- 3.00 kR gamma
rays + 1.50% EMS, AT
14
- 3.50 kR gamma rays + 0.50%
EMS, AT
15
- 3.50 kR gamma rays + 1.00% EMS, AT
16
-
3.50 kR gamma rays + 1.50% EMS. While imposing the
treatments, terminal cuttings were treated with Gamma rays
and EMS separately. But for combination of mutagenic
treatments, cuttings were initially treated with respective
EMS concentration and immediately then they were exposed
to Gamma radiation and planted in field in Randomized Block
Design (RBD) with 600 plants in each treatment.
Total number of branches (including primary,
secondary and tertiary branches) and leaves produced from
planted terminal cutting following mutagenic treatment was
represented as first vegetative generation and designated
as V
1
M
1
generation plants. Secondary shoots were
considered as the second vegetative generation. Secondary
shoots were obtained by cutting back the primary shoot and
planted for the study of V
2
M
1
generation. The mutants were
evaluated by adopting standard recommended cultural
practices (Hegde, 2001 and Rajamani, 2003) for crop
cultivation. The biometrical traits viz., plant height, number
of branches plant-1 and number of leaves plant-1 were
observed at 180 days after planting. Similarly, yield
parameters like length and girth of tuber and fresh tuber
yield plant-1 were also recorded. After the harvest of tubers,
the quality traits like forskolin (Mersinger et al,1988),
essential oil (A.S.T.A, 1960) and total alkaloids (Kokate et
al, 2001) were estimated by adopting standard procedures.
In V
2
M
1
generations, the genotypic correlation coefficients
and phenotypic correlation coefficient were estimated
according to Johnson et al (1955). The significance of the
genotypic correlation coefficients was tested by referring
to the standard table given by Snedecor and Cochran (1967).
Path coefficient analysis was carried out according to
Dewey and Lu (1959) by partitioning the genotypic
correlation into direct and indirect effects.
RESULTS AND DISCUSSION
The correlation coefficients between yield and its
components and inter correlations among various yield
attributes were estimated. In general, genotypic correlation
coefficients were of higher in magnitude than phenotypic
correlation indicting the lesser influence of environmental
factors. Being a complex trait, tuber yield is largely
influenced by many component characters. The relationship
between yield and its component characters is likely to vary
according to the genetic material used and environment
under which the material is evaluated as well as due to
interaction of these factors (Table 1 and 2).
The highest positive and significant genotypic
correlation of yield was observed with tuber girth (0.997)
and it was closely followed by number of leaves plant-1
(0.962) and number of tubers plant-1 (0.958). Other traits
exhibited positive and significant genotypic correlations with
yield are tuber length (0.934), plant height (0.906) and
number of branches plant-1 (0.847). While the characters
viz., forskolin (-0.782) and essential oil content (-0.167)
showed negative correlation with yield. Intercorrelation
showed that the plant height had positive and highly
significant association with number of branches plant-1,
number of leaves plant-1, number of tuber plant-1, tuber length
and tuber girth. Number of tuber plant-1 exhibited positive
and highly significant association with tuber length and tuber
girth. Each of these characters not only had positive
association with each other but also highly significant with
yield.
The yield exhibited positive and significant
phenotypic correlation with plant height, number of branches
plant-1, number of leaves plant-1, number of tuber plant-1,
tuber length and tuber girth. However, the forskolin and
essential oil showed negative correlation with yield. This
apparent negative correlation at genetic level would have
arisen from repulsion linkage of gene(s), controlling the direct
and indirect effects. Conversely, positive association was
due to the coupling phase of linkage. This is in agreement
with the earlier findings of Geetha and Prabhakaran (1987),
J. Hortl. Sci.
Vol. 4 (1): 63-67, 2009
Datta et al
65
Table 1. Effect of gamma rays (kR) and EMS on mean values for different traits in V
2
M
1
generation in Coleus
Treatment Biometrical traits at Yield trait Quality trait
180 Days from planting
Plant Number Number Number Length Girth Fresh Total Essential Forskolin
height of of of of of Tuber-yield alkaloid oil content (g)
(cm) branches leaves tubers tuber tuber plant-1(g) content(g) (ml)
plant-1 plant-1 plant-1 (cm) (cm)
AT1 Control 63.50 53.50 235.00 24.50 24.70 3.45 510.00 1.20 0.10 0.40
AT2 2.50 kR gamma rays 62.00 51.00 222.50 21.00 22.50 3.25 505.00 1.19 0.09 0.52
AT3 3.00 kR gamma rays 60.55 50.50 216.50 20.50 20.69 3.00 471.66 1.25 0.11 0.50
AT4 3.50 kR gamma rays 59.20 49.95 211.00 19.00 19.00 2.95 430.85 1.25 0.09 0.48
AT5 0.50 % EMS 58.00 49.00 229.50 23.50 23.50 3.30 497.55 1.19 0.10 0.42
AT6 1.00 % EMS 56.50 48.65 215.00 21.00 21.00 3.00 462.90 1.15 0.11 0.45
AT7 1.50 % EMS 55.00 48.05 203.00 18.00 19.80 2.85 433.20 1.10 0.09 0.40
AT8- 2.50 kR gamma rays 55.10 46.50 210.50 21.50 20.90 3.20 492.40 1.21 0.07 0.40
+ 0.50% EMS
AT9 2.50 kR gamma rays 53.60 45.00 200.00 18.50 19.10 3.18 477.60 1.20 0.18 0.45
+ 1.00% EMS
AT10 2.50 kR gamma rays 53.00 44.95 192.00 18.00 18.45 3.12 445.18 1.17 0.09 0.62
+ 1.50% EMS
AT11 3.00 kR gamma rays 52.50 44.00 182.00 16.50 17.65 3.00 420.00 1.14 0.12 0.46
+ 0.50% EMS
AT12 3.00 kR gamma rays 50.00 43.70 178.50 14.50 17.00 2.95 409.28 1.10 0.10 0.46
+ 1.00% EMS
AT13 3.00 kR gamma rays 49.00 42.80 160.00 13.00 16.20 2.65 389.30 0.96 0.09 0.55
+ 1.50 % EMS
AT14 3.50 kR gamma rays 45.50 41.20 148.00 12.00 15.00 2.45 375.50 1.10 0.10 0.50
+ 0.50% EMS
AT15 3.50 kR gamma rays 43.10 40.15 131.50 10.50 14.33 2.30 305.00 1.26 0.14 0.61
+ 1.00% EMS
AT16- 3.50 kR gamma rays 42.35 39.80 108.00 9.00 13.40 2.05 290.55 1.09 0.10 0.63
+ 1.50% EMS
Bhandari and Gupta (1991), Prabhakar et al (1994) and
Shanmugasundaram (1998). Correlation coefficients between
the characters revealed that those characters exerted positive
association among others are prone for improvement and
underlined the fact that one component character leads to
the concurrent improvement of the other component
characters. The present findings are concurrent with
Shanmugasundaram (1998) in turmeric and Kavitha (2005)
in coleus. The present information on strength and direction
of correlation of these component characters on tuber yield
and inter se association among them would be useful in
designing breeding programmes for yield improvement.
Correlation coefficient between any two characters
would not give a complete picture for a situation like yield,
which is controlled by several other traits, either directly or
indirectly. In such situations, path coefficient analysis
furnishes a means of measuring direct effect of each trait
as well as indirect effect via other characters on yield. So
information on direct and indirect effect on yield is important,
which is explicable by path analysis proposed by Wright
(1921) and illustrated by Dewey and Lu (1959). The
interrelationships of the component characters on yield
provide the likely consequences of their selection for
simultaneous improvement of desirable characters with
yield. The path analysis of component traits on yield of coleus
mutants showed positive direct effects through the
characters viz., plant height (0.979), number of leaves plant-
1 (0.422), number of tubers plant-1 (0.169), tuber length
(0.386), tuber girth (0.048) and essential oil content (0.008)
(Table 3). The direct effect was the highest for plant height
(0.979) followed by number of leaves plant-1 (0.422), while
the trait, number of branches plant-1 had the highest indirect
effect (-0.930). Since correlation of these characters with
yield is positive, preference should be given to these
characters in selection programme to isolate superior mutants
with genetic potential for improving yield. A similar line of
work was reported by Viswanathan et al (1993) in Abrus
precatorius and Srivastava and Chauhan (1998) in Bauhinia
variegata.
The direct effect was observed to be negative
through number of branches plant-1 (-0.930), total alkaloids
(-0.066) and forskolin content (-0.026). This vivid conflict
Statistics on Coleus mutants and yield
J. Hortl. Sci.
Vol. 4 (1): 63-67, 2009
66
between the correlation and path coefficient analysis arouse
largely from the fact that correlation simply measures the
mutual association without regard to causation, while path
specifies the relative importance of each causal factor. So
information on direct and indirect effect on yield is important
which is explicable only by means of path analysis. It is also
evident from the study that direct selection can be made on
tuber characters as they are true components relating to
yield and selection on these will be rewarding. The current
investigation resulted with the residual effect of 0.158
indicating precision on selection of component characters.
Most of the breeding programmes preferred with residual
effect lesser than one. It indicates that accuracy and
appropriate selection of component character for crop
improvement programme. This is supported by the earlier
works of Nandi et al (1992), Maurya et al (1998),
Shanmugasundaram (1998), Ushanandhinidevi (2004) in
turmeric and Kavitha (2005) in coleus. On a wholesome,
the weightage must be given to component characters
exhibiting positive attributes towards the fresh tuber yield
Table 2. Effect of gamma rays (kR) and EMS (per cent) on genotypic and phenotypic correlation coefficient in Coleus mutants in
V
2
M
1
generation
X
1
X
2
X
3
X
4
X
5
X
6
X
7
X
8
X
9
X
1 0
X
1
1 0.983** 0.958** 0.943** 0.934** 0.888** 0.415 -0.250 -0.697** 0.906**
1 0.983** 0.959** 0.929** 0.938** 0.852** 0.471 -0.140 -0.435 0.909**
X
2
1 0.928** 0.919** 0.930** 0.818** 0.418 -0.284 -0.688** 0.847**
1 0.931** 0.910** 0.933** 0.785** 0.465 -0.181 -0.445 0.854**
X
3
1 0.980** 0.955** 0.968** 0.406 -0.197 -0.805** 0.962**
1 0.972** 0.957** 0.901** 0.449 -0.108 -0.552* 0.963**
X
4
1 0.985** 0.981** 0.467 -0.192 -0.751** 0.958**
1 0.973** 0.852** 0.453 -0.152 -0.590* 0.951**
X
5
1 0.924** 0.381 -0.233 -0.779** 0.934**
1 0.875** 0.436 -0.130 -0.514* 0.937**
X
6
1 0.291 -0.178 -0.897** 0.997**
1 0.464 0.055 -0.328 0.919**
X
7
1 0.215 -0.247 0.321
1 0.331 0.046 0.370
X
8
1 0.008 -0.167
1 0.192 -0.085
X
9
1 -0.782**
1 -0.541*
X
10
1
1
# Upper values refers to genotypic correlation coefficient * Significant at 5 % level
# Lower values refers to phenotypic correlation coefficient ** Significant at 1 % level
X
1
: Plant height X
4
: Number of tubers plant-1
X
2
: Number of branches plant-1 X
5
: Length of tuber
X
3
: Number of leaves plant-1 X
6
: Girth of tuber
X
7
: Total alkaloids
X
8
: Essential oil
X
9
: Forskolin
Table 3. Effect of gamma rays (kR) and EMS (per cent) on path analysis in Coleus mutants in V
2
M
1
generation
X
1
X
2
X
3
X
4
X
5
X
6
X
7
X
8
X
9
Yield
X
1
0.979 0.318 0.338 -0.179 -0.386 -0.048 -0.043 -0.004 0.018 0.993
X
2
0.952 -0.930 0.392 0.155 0.359 0.040 -0.028 -0.002 0.018 0.956
X
3
0.968 -0.989 0.422 0.166 0.368 0.047 -0.027 -0.002 0.021 0.974
X
4
0.896 -0.991 0.414 0.169 0.380 0.048 -0.031 -0.001 0.019 0.903
X
5
0.944 -0.981 0.403 0.166 0.386 0.045 -0.025 -0.002 0.020 0.956
X
6
0.897 -0.882 0.409 0.166 0.356 0.048 -0.019 -0.001 0.023 0.997
X
7
0.419 -0.451 0.171 0.079 0.147 0.014 -0.066 0.002 0.006 0.321
X
8
-0.252 0.306 -0.083 -0.032 -0.090 -0.009 -0.014 0.008 0.000 -0.166
X
9
-0.704 0.742 -0.340 -0.127 -0.301 -0.043 0.016 0.000 -0.026 -0.783
# Residual effect: 0.158 Diagonal element - Direct effects
X
1
: Plant height
X
2
: Number of branches plant-1
X
3
: Number of leaves plant-1
X
1 0
: Fresh tuber yield plant-1
X
7
: Total alkaloid content
X
8
: Essential oil content
X
9
: Forskolin content
X
4
: Number of tubers plant-1
X
5
: Length of tuber
X
6
: Girth of tuber
X
1 0
: Fresh tuber yield plant-1
J. Hortl. Sci.
Vol. 4 (1): 63-67, 2009
Datta et al
67
of coleus. However, certain traits with negative attributes
are also chosen for getting improved quality i.e., forskolin
content without much inhibition on fresh tuber plant-1.
REFERENCES
A.S.T.A. 1960. Official analytical methods of the American
spice trade association, New York, pp. 41-42
Bhandari, M.M. and Gupta, G.S. 1991. Association analysis
of Opium poppy Linn. Int’l. J. Trop. Agri., 9 (1):
42-44
Bhat, S.V., Bajwa, B.S., Dornauer, H., DeSouza, N.J. and
Fehlilhaber, H.W. 1977. Structure and
stereochemistry of new labdane diterpenoid from
Coleus forskohlii. Tetrahedron Lett., 19:1669-1672
Dewey, D.R. and Lu, K.H. 1959. A correlation and path
coefficient analysis of components of crested wheat
grass seed production. Agron. J., 51:515-518
Geetha, V. and Prabhakaran, P.V. 1987. Genotypic
variability, correlation and path coefficient analysis
in turmeric. Agril. Res. J. Kerala, 25:249-254
Hegde, L. 2001. Crop improvement in Coleus forskohlii
Briq. In : National seminar on transfer of technology
of medicinal and aromatic crops, held on 20-22
February, Bangalore, pp. 92-96
Johnson, H.W., Robinson, H.F. and Comstock, R.E. 1955.
Estimation of genetic variability in soybean. Agron.
J., 47:314-318
Kavitha, C. 2005. Genetic diversity studies in Coleus
forskohlii Briq. Ph.D thesis, TNAU, Coimbatore
Kokate, C.K., Prohit, A.P. and Gokhale, S.B. 2001.
Pharmacognosy, Nirali Prakasam Publ., Warangal,
India.
Maurya, K.R., Kumar, R., De, N. and Singh, S.N. 1998.
Correlation and path analysis performance studies of
some turmeric (Curcuma domestica Val.) genotypes.
In: National Seminar on recent development in spices
production technology, Bihar Agricultural College,
Sabour. pp. 11-12
Mersinger, R., Donauer, H. and Rainhard, E. 1988. Formation
of forskolin by suspension cultures of Coleus
forskohlii. Planta Med., 54:200-204
Nandi, A., Lenka, D. and Singh, D.N. 1992. Path analysis
in turmeric. Ind. Cocoa, Arecanut and Spices J.,
17:54-55
Prabhakar, M., Singh, J. and Srivastava, L.J. 1994. Genetic
variation and correlation studies for some important
characters associated with solasodine yield in
Solanum khasianum Clarke. Ind. J. Forestry,
17:26-31
Rajamani, K. 2003. Cultivation technology of Coleus
forskohlii. In: State level sem. Medicinal Plants, held
at Trichy, pp: 53-58
Rupp, R.H., De Souza, N.J. and Dohadwalla, A.N. 1986.
In: Proceedings of the International Symposium on
Forskolin: Its chemical, biological and medical
potential. Hoechst India Ltd., Bombay, pp. 19-30
Shanmugasundaram, K.A. 1998. Evaluation and selection
for certain quantitative and qualitative characters in
turmeric (Curcuma domestica Val.). M.Sc. (Ag.)
thesis, TNAU, Coimbatore
Snedecor, G.W. and Cochran, C.W.G. 1967. Statistical methods.
The Iowa State University Press, Iowa, U.S.A.
Srivastava, A. and Chauhan, K.C. 1998. Path coefficient
analysis between shoot dry weight and other
characters in Bauhinia variegata Linn. progeny. J.
Tree Sci., 15:111-113
Ushanandhinidevi, H. 2004. Induction of mutagenesis in
turmeric (Curcuma longa L.) through gamma rays
for variability and quality improvement. Ph.D thesis,
TNAU, Coimbatore
Vishwanathan, T.V., Sunil, K.P. and Sujatha, R. 1993. Path
analysis for lethality in gamma irradiated population
of Indian liquorice (Abrus precatorius L.) South Ind.
Hort., 41:101-105
Wright, S. 1921. Correlation and causation. J. Agril. Res.,
20:557-585
(MS Received 30 April 2008, revised 12 February 2009)
Statistics on Coleus mutants and yield
J. Hortl. Sci.
Vol. 4 (1): 63-67, 2009