You are on page 1of 4


In this case we would not nec- Hooked Trichomes: A Physical Plant Barrier
essarily expect to get very good agree-
ment between calculations and experi- to a Major Agricultural Pest
mental results, but those calculations
could be used as an upper limit for the ef- Abstract. Hooked epidermal appendages (trichomes) on leaves offield bean culti-
fect of nonsphericity on scattering char- vars effectively capture nymph and adult leafhoppers. Frequency of capture and cap-
acteristics of interest. The result of such ture mortality are highly correlated with trichome density. Hooked trichomes in-
calculations, as shown in Fig. 3, can be serted at angles less than 300 are ineffective in capture.
useful in many applications. For ex-
ample, in deducing the physical proper- Breeding crop plants for genetic resist- chomes of P. vulgaris and mortality of E.
ties of atmospheric aerosols or cosmic ance to insect pests is an ecologically fabae and report the basis for intra-
dust particles from scattering data, one compatible alternative to the use of specific and interspecific variation in the
can use first the Mie calculation assum- chemical pesticides. Selection of resist- effectiveness of this defense mechanism.
ing spherical particles and then the modi- ant cultivars can be accelerated if specif- During studies of feeding damage of E.
fied Mie calculations with n1 = 1. Re- ic resistance mechanisms are identified. fabae on field beans, we observed leaf-
gardless of the shape of aerosol particles, Although the role of secondary chem- hopper nymphs and adults clinging to
the correct values of physical properties icals in plant resistance to herbivore at- leaves of certain cultivars. Using a dis-
should be somewhere between the two tack has received much emphasis (1), secting microscope at 60X, we observed
sets of results, as in Fig. 3. considerably less is known about physi- leafhoppers physically impaled on leaf
We again emphasize that our approxi- cal defensive barriers in plants. Recent hairs. Closer examination revealed large

Downloaded from on November 19, 2016

mation should be used only for en- review articles (2) have brought attention numbers of hooked trichomes on the un-
sembles of randomly oriented and arbi- to the involvement of glandular and non- dersurface of the leaves. These special-
trarily shaped nonspherical particles. glandular plant hairs (trichomes) in in- ized hairs ranged from 0.06 to 0. I1 mm in
Whenever all particles are oriented in a sect resistance. length; hairs over the leaf veins were
definite direction or whenever the scat- Numerous attempts have been made slightly longer. As leafhoppers are very
tering occurs on a single particle, specif- to establish a statistical correlation be- active, we concluded that the insects had
ic effects depend on a particle's exact tween the degree of crop pubescence and become entangled in the hairs as they
shape; since these effects are not includ- resistance to leafhoppers of the genus moved rapidly over the leaf surface.
ed in our approximation, it should not be Empoasca (Walsh) (3), major worldwide To document the various methods of
used. In addition, our intercomparisons pests of cotton, soybeans, hIfalfa, clo- capture, we examined captured leafhop-
ver, field beans, and potatoes. In North pers on bean leaves by scanning electron
were made for particles in the Mie size
range x < 30. This adequately covers the America Empoasca fabae (Harris), the microscopy (9). Both nymphs and adults
size range, for example, of the normal potato leafhopper, is a serious pest of were found to be impaled through the in-
background aerosol particles in the field beans, Phaseolus vulgaris L. In tersegmental membranes of the abdomen
earth's atmosphere; further studies will South America and Central America a by hooked trichomes (Fig. 1, a and b), to
be required before the procedure can be closely related species, Empoasca our knowledge a previously unreported
applied to larger sizes. kraemeri Ross and Moore, is a primary phenomenon. Insects were also captured
PETR CHY'LEK limiting factor in the production of this vi- by trichomes impaled in the tarsal seg-
Department ofGeosciences, tal protein crop (4). Although there have ment of the leg or entangled in the tarsal
Purdue University, been anecdotal reports of arthropod cap- claws, as reported for other insect spe-
West Lafayette, Indiana 47907, and ture by hooked trichomes of P. vulgaris cies (6, 7). Figure Ic illustrates yet anoth-
National Center for Atmospheric (5-7), the specific impact of pubescence er form of capture, by a trichome embed-
Research, Boulder, Colorado 80303 in different cultivars of field beans on the ded in the membrane between the tibia
G. W. GRAMS population biology of leafhopper pests and the tarsus.
National Center for Atmospheric does not appear to have been determined Leafhoppers captured by the tarsus or
Research (8). In this report, we describe a direct tibia terminate feeding and struggle to
R. G. PINNICK relationship between the hooked tri- free themselves from the trichomes, fre-
Atmospheric Sciences Laboratory,
U.S. Army Electronics Command,
White Sands Missile Range, Table 1. Relationship between density of hooked trichomes and capture mortality of leafhopper
New Mexico 88002 nymphs caged on leaves of field beans and lima beans. Means followed by the same letter within
a column are not significantly different at P = .05 by Duncan's multiple range test.
References and Notes
1. A. C. Holland and J. S. Draper, Appl. Opt. 6, trichome Capture
511 (1967); A. C. Holland and J. Gagne, ibid. 9, Capture
1113(1970). Cultivar Node sLrface density(er mortality
2. J. E. Hansen and L. D. Travis, Space Sci. Rev. (No. per a) (%
16, 527(1974). cm2)
3. H. C. van de Hulst, Light Scattering by Small
Particles (Wiley, New York, 1957), chap. 17. Phaseolus vulgaris Terminal Lower 14,241 a 59.8 a 36.8 a
4. P. Chylek,J. Opt. Soc. Am. 66, 285 (1976).
5. ,ibid. 65, 1316 (1975). 'California Light 4 Lower 1,955 b 50.0 b 31.5 ab
6. G. W. Grams, A. J. Dascher, C. M. Wyman, Red Kidney' 2 Lower 1,593 b 36.4 c 26.8 b
Opt. Eng. 14, 80 (1975). 2 Upper 244 e 5.7 ef 2.6 d
7. G. Herdan, Small Particle Statistics (Butter-
worths, London, 1960), chap. 6. Phaseolus vulgaris 4 Lower 435 d 16.9 d 12.4 c
8. The National Center for Atmospheric Research 'Brasil 343' 2 Lower 362 de 13.2 de 7.6 cd
is sponsored by the National Science Founda-
tion. P.C. was partially supported by a grant 2 Upper llf 0.5f 0.Od
from the Atmospheric Science Section of the Phaseolus lunatus 2 Lower 841 c 4.5 ef 2.0 d
National Science Foundation.
'Henderson Bush Lima'
31 March 1976; revised 2 June 1976
482 SCIENCE, VOL. 193
quently becoming impaled on other tri- a constant temperature of 27C and a 16- ined and the numbers of captured and
chomes during the struggle. If unable to hour photoperiod. dead nymphs determined. A nymph was
escape, the insect usually dies from dehy- Thirty-one days after seeding, first and recorded as captured only if a trichome
dration and starvation. We observed cast second instar nymphs of E. fabae were was actually observed entering the body
skins attached to the leaf by a trichome, aspirated from stock plants of broad wall or leg.
indicating that some nymphs escape by bean, Vicia faba L.; they were anesthe- The density of hooked trichomes is
the process of molting. Abdominal cap- tized with CO2 and confined to leaf sur- highly correlated with capture frequency
ture is a more lethal form of plant de- faces by using small cages made of Plexi- and nymphal mortality of E. fabae on the
fense. As the insect struggles to escape, glas tubing (inner diameter, 23 mm) (12, two field bean cultivars (Table 1). The
the abdominal wall frequently ruptures 13). Ten nymphs were placed in each lower leaf surfaces have higher densities
and death soon follows from loss of he- cage and two cages were used per leaflet, than the upper surfaces. As leafhoppers
molymph or from starvation and dehy- for a total of 20 nymphs per treatment feed almost exclusively on the underside
dration. (cultivar and leaflet) for each replication. of the leaves, the trichomes are ideally lo-
To relate trichome density to capture After caging and infestation, a leaf sam- cated for defensive purposes. Within cul-
efficiency, we examined the field bean ple was taken from another leaflet of the tivars, mature leaves on the lower nodes
cultivars P. vulgaris 'California Light same trifoliolate leaf, cleared of all pig- have fewer hooked trichomes than ma-
Red Kidney' and P. vulgaris 'Brasil 343' mentation in 70 percent ethanol, and ex- ture leaves on the upper nodes. This dif-
with respectively high and low densities of amined at lOOx underacompound micro- ference is consistent with increased cap-
hooked trichomes. A lima bean, Phase- scope. Ten counts of the number of ture and capture mortality on the upper
olus lunatus 'Henderson Bush,' was also hooked trichomes in a 1.56-mm2 field of nodes. Trichome density and capture

Downloaded from on November 19, 2016

included, as this species is susceptible to the leaf surface were made randomly to mortality were at a maximum on the
damage by leafhoppers (10). The bean provide an average hooked trichome den- unexpanded terminal leaves. The num-
plants were raised in a controlled-en- sity for the whole leaf. Twenty-four hours ber of trichomes is fixed early in leaf de-
vironment glasshouse (11), maintained at after infestation, the leaves were exam- velopment (14); thus, the density of


Fig. 1. (a) Trichome inserted in abdomen of a leafhopper nymph (x 700). (b) Trichome embedded in posterior of abdomen (x 700). (c) Tn-
chome embedded in membranous tissue between leg segments (x 350). (d) Procumbent hooked trichomes of the lima bean cultivar Henderson
Bush(x 350).
6 AUGUST 1976 483
hooked trichomes decreases as the leaf a
References and Notes
expands. This provides the greatest in- 1. R. H. Whittaker and P. P. Feeny, Science 171,
sect protection for the tender and succu- 757 (1971).
40 -
2. D. A. Levin, Q. Rev. Biol. 48, 3 (1973); H. B.
lent plant tissues. Johnson, Bot. Rev. 41, 233 (1975); J. A. Web-
ster, Mich. Agric. Exp. Stn. Misc. Publ. 1297
Intraspecific variation in nymphal mor- aL)
tality was also accounted for by differ- 3. H. W. Johnson and E. A. Hollowell, J. Agric.
20 - Res. 51, 371 (1935); F. R. Parnell, H. E. King,
ences in trichome density. When cn. D. F. Ruston, Bull. Entomol. Res. 39, 539
nymphs were confined to upper leaf sur- I (1949); N. L. Tailor, Agron. J. 48, 78 (1956); D.
B. Broersma, R. L. Bernard, W. H. Luckmann,
faces (low trichome density), total mor- 0 J. Econ. Entomol. 65, 78 (1972).
tality for the two field bean cultivars was 2000
4. G. Wilde and A. van Schoonhoven, Environ.
0 1000 14000 Entomol. 5, 251 (1976).
the same. This suggests that the two cul- Number of hooked trichomes (cm2) 5. F. W. Poos and F. F. Smith, J. Econ. Entomol.
24, 361 (1931); H. H. Richardson (6); H. J.
tivars, excluding trichome effects, are Fig. 2. Relationship between hooked tri- deFluiter and G. W. Ankersmit, Tijdschr. Plant-
equally suitable as leafhopper hosts. For chome density and capture frequency of leaf- enziekten 54, 1 (1948); B. Johnson (7), L. E.
Gilbert [Science 172, 585 (1972)] has reported a
nymphs confined to the lower leaf sur- hoppers on field beans. similar phenomenon for Passiflora adenopoda
face, however, capture mortality was ap- with heliconiine larvae.
6. H. H. Richardson, J. Econ. Entomol. 36, 543
proximately three times greater on the (1943).
cultivar with the higher density of ularly if this defense mechanism can be 7. B. Johnson,Bull. Entomol. Res. 44, 779(1953).
8. D. A. Wolfenbarger and J. P. Sleesman, J.
hooked trichomes. combined with other plant factors medi- Econ. Entomol. 54, 845 (1961).
9. The insect specimens were coated with PdAu,
Overall, nymphal capture and capture ating resistance to egg laying and growth without prior dehydration, and examined with
mortality are highly correlated (r = .99, of leafhoppers. an Advanced Metal Research model 900 scan-
ning electron microscope. Photomicrographs

Downloaded from on November 19, 2016

P < .01) with hooked trichome density The role and importance of trichomes were made using Polaroid type 105 positive-
for the field bean cultivars (Fig. 2). Cap- in resistance to crop pests has frequently negative film.
10. C. J. Eckenrode and L. P. Ditman, J. Econ.
ture frequency increases rapidly with tri- been overlooked. In part, this may have Entomol. 56, 551 (1963).
chome density, but appears to level off 11. The three cultivars were seeded in peat-ver-
been due to the inability of earlier work- miculite soil mix (supplemented with Osmocote
beyond a density of 2000 trichomes per ers to consistently correlate total plant slow-release fertilizer), using clay pots 7 inches
in diameter. Ten pots of each cultivar were
square centimeter. We suggest that the pubescence with such general parame- arranged in a randomized complete block design
trichomes are so close together on termi- ters as insect infestation level or crop in a controlled-environment glasshouse; natural
light was supplemented by Metalarc illumina-
nal leaves that the nymphs are removed yield. As our experience with E. fabae tion.
from most direct body contact with the on field beans indicates, it is important to 12. W. M. Tingey, V. E. Gracen, J. M. Scriber,
N. Y. Food Life Sci. Q. 8, 3 (1975).
leaf surface. Thus, very high densities of identify the biological impact,on the pest 13. Several methods for confining leafhopper
hooked trichomes could reduce capture of a specific trichome type. Other fac- nymphs on leaves were compared. Although
capture frequency varied slightly among differ-
and decrease the efficiency of this plant tors, including trichome density and ent methods, nymphal capture was consistently
correlated with trichome density.
defense mechanism. Although removed angle of insertion, must also be consid- 14. K. Esau, Plant Anatomy (Wiley, New York,
from direct contact with the leaf surface, ered. This knowledge should permit 1965), p. 167.
15. Measurements of angle of insertion were made
these nymphs could continue to feed, as more effective selection of insect resist- with fresh material as well as leaf sections
their stylets are long enough to reach the ant cultivars. cleared in alcohol.
16. We are indebted to S. Temple and D. H. Wal-
leaf veins. ERIC A. PILLEMER lace for providing seed and to T. Eisner, P. P.
The percentage of caged leafhopper Department of Plant Breeding, Feeny, V. E. Gracen, and D. J. Paolillo for their
critical reviews of the manuscript. This work
nymphs captured by the lima bean culti- Cornell University, was supported, in part, by a Rockefeller Founda-
tion grant to D. H. Wallace. This is a publication
var was much less than the percentage Ithaca, New York 14853 of Cornell University Agricultural Experiment
captured by either of the field bean culti- WARD M. TINGEY Station, New York State College of Agriculture
and Life Sciences, a Statutory College of the
vars (Table 1). This interspecific varia- Department of Entomology, State University of New York.
tion is not explained by differential tri- Cornell University 29 March 1976; revised 24 May 1976
chome density, as lima bean leaves have
moderately large densities of hooked tri-
chomes. But the hooked hairs of P. luna-
tus are oriented at an angle of 100 to 300
(15) (Fig. Id), whereas the hairs of P. vul- Fog Catchment Sand Trenches Constructed by
garis are relatively erect. Apparently Tenebrionid Beetles, Lepidochora, from the Namib Desert
these procumbent trichomes are oriented
at such a small angle that the nymphs Abstract. Three species of coastal Namib Desert tenebrionid beetles (Lepido-
rarely come in contact with them. Thus chora) build trenches on desert sand dunes. Trenches are constructed perpendicular
the observed interspecific variability in to fog winds and concentrate moisture during fogs. The beetles return along the
nymphal capture for this cultivar is re- ridges of the trenches extracting waterfrom them. The water content ofa population
lated to angle of trichome insertion rath- of these beetles increased by 13.9 percent during one fog.
er than to density.
Clearly, hooked trichomes cause in- Field observations of three nocturnal al locomotory mode, however, is to walk
creased mortality of E. fabae on field beetle species, Lepidochora discoidalis, over the surface without displacing sand
beans. This reduction in nymphal infesta- L. porti, and L. kahani were made during (1). The ridges trap fog water, which is
tion has dramatic implications for vari- early summer (October 1975 through Jan- then taken up by the returning beetles.
etal plant resistance, as nymphal infesta- uary 1976), in the coastal African Namib During the early part of fogs, the
tion density is generally correlated with Desert dunes. These beetles establish beetles emerge on the barren sand dunes
the severity of leafhopper damage (8). trenches on the vegetationless surface of and move in a straight line, laying out
Thus, hooked trichome density may be a the sand dunes on mornings before and trenches as long as a meter or more.
valuable selection criterion in applied during advective fogs. Trenching creates The trenches become increasingly con-
breeding programs for field beans, partic- two parallel sand ridges (Fig. 1). The usu- spicuous as the parallel ridges trap blow-
484 SCIENCE, VOL. 193
Hooked Trichomes: A Physical Plant Barrier to a Major
Agricultural Pest
Science 193 (4252), 482-484. [doi:

Editor's Summary

Downloaded from on November 19, 2016

This copy is for your personal, non-commercial use only.

Article Tools Visit the online version of this article to access the personalization and
article tools:

Permissions Obtain information about reproducing this article:

Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last
week in December, by the American Association for the Advancement of Science, 1200 New
York Avenue NW, Washington, DC 20005. Copyright 2016 by the American Association for
the Advancement of Science; all rights reserved. The title Science is a registered trademark of