ABSTRACT
Suction capillary electrodes were employed to study the response of the
osphradial nerve to electrical and chemical stimulation in the limpet,
Colisella limatula. The osphradial nerve was filled with CoCl to identify
individual cell bodies and axons along the posterior edge of the osphradium.)
Chemo-sensitive paddle cilia on the surface of the osphradium were not
abundant, if present at all, when observed under the scanning electron
microscope. The organ proved to be sensitive to the application of IM NaCl,
IM Licl, and 0.5M Na-Gluconate + ASW; it gave a heightened response both in
frequency and amplitude of action potentials observed passing along the
osphradial nerve (O.N.). A second effect was also observed: the organ was
sensitive, on a smaller scale, to 0.6M Sucrose +ASW, and 0.5M Nacl + ASW.
The application of these substances resulted in smaller increases in
amplitude and frequency, but still visibly greater than that of natural sea
water (NSW) which had no effect. Ineffective substances for osphradial
stimulation were 0.5M N-Methyl-Glucamine-Cl, light, Pisaster ochraceus tube
feet, dacron line, and distilled water.
INTRODUCTION
The function of the gastropod osphradium is, in general, poorly
understood. Based on morphological studies, the osphradium in many
prosobranchs has been identified as chemosensory (Haszprunar, 1985 A).
The only physiological work has been carried out on the opistobranch.
Aplysia, where recordings were made from central nervous system cell bodies
which presumably were targets for axons coming from the osphradium via the
O.N. In these studies osmotic, mechano, and chemo all proved successful
stimulants (Stinnakre & Tauc. 1968) (Jahan-Parwar, et-al, 1969). A body of
behavioral literature also exists and supports the idea that the osphradium
in the predatory gastropod Biomphalcria glabrata serves as a food detector
(Townsend. 1973).
The osphradium of prosobranch limpets has not been previously
investigated physiologically. It is unlikely that it would serve as a
mechanoreceptor or sediment detector because limpets are found exclusively on
rocks. It is also unlikely that it is a food detector in this
non-carnivorous browsing gastropod. Therefore, the project attempted to
detect an osmo, or chemo receptor in the limpet, Colisella limatula. These
animals are subjected to osmotic stresses in their intertidal habitats.
The approach taken was to record electrical activity in the osphradial
nerve following stimulation of the organ's sensory epithelium.
These experiments represent the first such successful recording from the
O.N. of any gastropod.
MATERIALS AND METHODS
Specimins of Colisella limatula were collected from the Great Tide Pool
in Pacific Grove, California. Limpets were maintained in holding tanks in
running natural sea water at 16°0.
Dissection Technique
To expose the osphradial nerve, an animal was deshelled with a scalpel
blade, placed in a slygard bottomed dish filled with sea water, and
restrained using insect pins. An incision was made in the mantle tissue
along the dorsal midline running posteriorly from the head of the organism.
The two flaps of mantle were then gently pulled back and the connective
tissue cut away from the mantle- exposing the osphradial nerve. The left
osphradium was used in all experiments because it was more easily isolated
and recorded from than was the right. All tissue posterior to the O.N. was
removed to reduce the possibility of muscular contraction during stimulation.
All dissections and experimentations were carried out in natural sea water.
Experimental Set Up
An electrical stimulating electrode was constructed of fine Pt wires
embedded in silicon glue to expose only the ends of the wires. The electrode
was placed on different areas on the left osphradium to map sensitivity and
confirm that the recording electrode was properly positioned on the O.N.
Rectangular pulses of 0.3 ms duration and 8-40 volts were employed.
Recording suction electrodes made from glass hematocrit tubes were pulled on
a micro-pipette puller and broken to a diameter approximately equal to that
of the nerve (150 um). The recording electrode measured the voltage
difference between two Ag:AgCl wires: one placed inside the syringe, and the
other wrapped around the glass tip— careful insulation of this wire with
Q-dope, except for the tip, was important to minimize the stimulus artifact.
The recording electrode was filled with NSW. Recordings were amplified 1000
times by a Tektronix FM122, filtered at 1000Hz (high-pass) and 8Hz
(low-pass), and displayed on a storage oscilloscope and Gould 220 chart
recorder.
Positioning of the electrode was on the osphradial nerve before the
split leading to the left and right pleural ganglia (see Fig. 1). Slight
suction was applied to the nerve until either spontaneous activity was
observed or the nerve was sucked up into the electrode, forming an omega
shape. The nerve was then slightly raised from the body to reduce the
possibility of mechanical noise interferring with the recording of action
potentials. In all cases, before testing for chemically stimulated nerve
activity, positive responses were obtained with electrical stimulation of the
osphradium or of the nerve itself.
Chemical stimulation of the nerve originally consisted of introducing a
hand micropipettor into the bath, slowly releasing 250 ul of solution onto
the organ, and then removing the pipettor from the bath. To wash the organ
with a new substance the process was repeated.
It was felt that this method of application left too many uncontrolled
variables and often generated unacceptable electrical and mechanical noise.
The application method was then changed to a three barreled solution
distributor mounted on a micromanipulator which could be turned on and off at
will, allowing one solution to be dontindousky fidetng. Sach bagcel
cioposasie pipettor tip) was fed from Intramedic Polyethylene tubing attached
to an individually valved 60ml syringe. Flow from each of the barrels could
be started or stopped with minimal mechanical or electrical disturbance, Flow
rate was approximately one drop per second into a dish containing 50 ml of
NSW.
Three barrels for bathing were used so that one could be stopped and
another started without stopping the general flow into the medium. During
each experiment one barrel was filled with NSW. One barrel was filled with a
solution known to elicit a positive response (e.g. 1H Nacl) as a check of
organ viability. The third barrel was filled with an experimental solution
to be tested and compared.
Control solutions of NSW and ASW were allowed to flow on the organ for 2
to 4 minutes to ensure equiliberation. Experimental and the positive
response check solutions flowed between 1 and 5 minutes. At the conclusion
of each experiment the nerve was electrically stimulated to reconfirm that
the nerve was still in the recording electrode.
Experimental Solutions
Hypertonic IM Nacl was used to increase Na, Cl, and osmolarity. O.SM
Na-Cluconate i ASW was used to vary the Na while keeping Cl constant, while
raising the osmolarity. O.5M N-Methyl-Clucamine-Cl 4 ASW was used to raise
the number of Cl ions in the solution, keeping the Na normal, and raising the
osmolarity. O.6ti Sucrose + ASW was used to raise the osmolarity while
keeping the Na and Cl concentrations normal. A lM LiCl solution was prepared
to test if the osphradium was a Na specific receptor, or sensitive to high
osmolarity (see table 1 for actual concentrations).
Scanning Electron Microscope Preparation
specimens were prepared for scanning electron microscopy by fixation in
a 12 glutaraldehyde, 12 para-formaldehyde, 0.22 Na Cacodylate, 857 sea water
solution; dehydration was in acetone. Critical point drying with Co was
employed to avoid the grape-to-raisin effect.
CoCla Filling of the Nerve
The preparation of the animal for the cobalt fill was done in a cold
room (4'C) to keep the specimen fresh. The nerve was cut and sucked up into
a capillary electrode filled with 350mM CoCla buffered to PI 7.0 with Tris.
This solution was slightly hypotonic (830 mosmols). The electrode was
attached to a 6 volt battery through a 10M ohm resistor, and a current passed
between the electrode and an external Ag:AgCl wire. The preparation was
allowed to sit for 9-18 hours. The electrode was then removed and the
specimen was washed in NSW for four minutes. The cobalt was then
precipitated with a 1-5% solution of ammonium sulfide, afterwhich, it was
placed in 102 formalin + NSW overnight to fix. Dehydration was with ethanol.
Specimens were mounted in Permount and viewed under a compound microscope,
SULTS
Anatomy of the Osphradium and Osphradial Nerve
Filling nerves with CoCly allowed positive identification of the nerve
that extends to the left osphradium, but no part of the actual osphradial
organ was stained with cobalt. Individually stained cell bodies (glou dia.
were found lying within the mantle tissue along the posterior edge of the
organ (see Fig. 2). Single filled axons eminating from these cell bodies
were seen transversing the mantle beneath the organ in the direction of the
nerve trunk. Positive identification of filled cell bodies or axons in other
areas (i.e. along the medial edge of the organ) was not made. Presumably the
cell bodies which were stained are part of the osphradial ganglion often
referred to in the literature, and axons of these cells pass out via the O.N.
to the more centralized ganglia.
The use of the scanning electron microscope to identify chemo-receptors
yielded no definite results. The organ appears to be covered with fissures
and cracks running throughout (Fig 34). Under greater magnification (Fig 3B)
it is evident that the entire surface of the osphradium is composed of
microvilli. Chemoreceptive paddle cilia were not abundant, if present at
all, in the Colisella limatula osphradia examined.
Electrical Stimulation of the Osphradium
Osphradial nerve activity due to electrical stimulation of the
osphradium was routinely observed. Electrically stimulated action potentials
occurring with the lowest threshold, and presumably reflecting single unit
activity (Fig 44) and spontaneously occurring action potentials (Fig 4B) were
roughly similar in magnitude and time course. The lowest threshold responses
could only be obtained from areas of the organ which were much more sensitive
to electrical stimulation than others. These areas were the posterior and
medial edges of the organ (Fig 2). These areas probably correspond to where
the cell bodies of the O.N. axons lie. Stimulation of areas of the maatle
adjacent to the organ did not lead to significan activity in the O.N. (Fig
5).
Osphradial Nerve Activity Due to Stimulation by Hypertonic Nacl
Activity in the O.N. was also routinely stimulated by application of
250ul of 1M Nacl. Figure 6 shows the result of applying 1M Nacl directly to
the sensory epithelium of the left osphradium. Both frequency and amplitude
increased in the spikes recorded from the O.N. (see Fig 7). It can be seen
that the time course of spikes during 1M Nacl exposure (Fig 6) was very
similar to those of the control solution in figure 4B.
The increased spike amplitude and frequency were maintained until the
osphradium was washed with NSW (in Fig 7 260sec). Once washed, frequency
and amplitude soon returned to normal. Application of IM Nacl to the
anterior and posterior ends of the mantle or foot of the limpet did not vield
results. Similarly, application to the cephalic tenticles did not cause
excitation of units in the O.N.
Several problems with the hand pipettor method used to obtain the
results in figure 7 stimulated development of the three barreled method of
applying solutions (see methods). Mechanical and electrical noise produced
by solution changes was thereby minimized. All of the remaining experiments
to be described were carried out using this three barreled method. Figure 84
shows that turning off and on the sea water control flow caused little change
in O.N. activity. The application of artificial sea water also caused no
stimulation (not illus.). Bathing the organ in M NaCl yielded a
characteristic effect upon every application. Both spike amplitude and
frequency increased markedly in IM NaCl and returned to normal when NSH was
reintroduced (Fig 80).
Application of 1M Nacl, besides causing an increase in the frequency and
amplitude of O.N. spikes, also apparently resulted in a heightened ammount of
baseline noise (as in Fig. 8A & B). This enlargement was only seen when
solutions led to an obvious increase in spikes of ) 2-3uv, and promptly
dissapeared when the control solution (NSW) was reapplied and the spikes
decreased in'frequency (Fig 8B & C). Increased baseline could be due to
either a large increase in small spikes (2-3 uv, or coincidental
amplification of background electrical noise, but there is no apparent reason
for the latter possibility.
Frequency of O.N. Discharge in Response to lM Nacl
Figure 9 shows the frequency of spikes (in 2 sec bins), which are
larger than 2uv, through time. Application of 1M Nacl solution causes a
large increase in the frequency of such spikes. The time to reach peak
effect and return to the control level, following readmission of NSW, is
approximately four seconds. This trend of heightened frequency was typical
for all experiments using 1M Nacl.
Analysis of Spike Amplitude Distributions
Because of the uncertain origin of the increased baseline noise
accompanying the Nacl stimulated spikes of large amplitude and high
frequency, the distribution of spike amplitudes was initially analyzed by
measuring spikes protruding from the baseline noise, thus assuming the
increase in the baseline was due to amplification of ambient electrical noise
(Fig 7 inset). As discussed above, however, a more reasonable source of the
apparent 'noise' is a large increase in small spike activity. Spike
amplitudes were therefore measured from the middle of the baseline as
indicated in figure 7 inset.
Using the first method of calculation it can be seen (Fig 104) that the
application of 1M Nacl to the osphradium causes a great increase in the
number of mid-sized spikes 3-6uv, a smaller increase of larger spikes (buy).
and apparent decrease in spikes (3uv. Due to the heightened baseline
problem, however, measurments of spikes 43uv are not very reliable in IM
Nacl. When the second method described above to measure spike amplitude is
used, it basically ignores spikes (3uv. This analysis (Fig 10B) shows that
the 1M Nacl distribution probably follows the NSW curve at the smaller
amplitudes, but deviates sharply from it above 4uv. The primary effect of 1M
Nacl can be described as a great increase in spikes Youv which are very rare
in NSW. It is important to note that the qualitative effect of IM Nacl is
not dependent on the method of analysis.
The effect of 0.5M Nacl + ASW on the spike amplitude distribution is
also plotted in figures 10 A & B. Using the first method of analysis, the
effect is a marginal increase in the 3-6uv spikes (Fig 10A). Following the
second method, however, a dramatic increase in the number of mid-sized spikes
becomes evident. Few very large spikes appear in 0.5M Nacl + ASW. Again.
10
the qualitative effect of 0.5M Nacl + ASW is the same in figures 10 A & B.
but for reasons discussed above, the picture given in figure 10B is taken to
be the more realistic one.
Analysis of all further amplitude distributions were carried out using
the second method of spike measurment. All calculations were done from four
second blocks occurring four seconds after changing solutions (to allow
equiliberation, see figure 9). Numbers smaller than 2-3uv are too ambiguous
to count, because they are lost in the heightened baseline noise accompaning
increased activity.
Experimental Variations of Na, Cl, and Osmolarity
Because 1M Nacl and 0.5M Nacl + ASW both had definite effects on
activity in the O.N., it was believed that some aspect associated with the
increased NaCl content of these media evoked the response. Variations of
concentrations of Na, Cl, and osmolarity were tested. O.5M Na-Gluconate +
ASW (see table I) was used to double the concentration of Na, and to increase
osmolarity, while keeping Cl at its normal level. O.5M N-Methyl-Glucamine-Cl
+ ASW was used to increase the concentration of Cl and osmolarity, while
keeping Na normal. Finally, O.6M Sucrose + ASW was used to increase
osmolarity alone, while keeping both Na and Cl concentrations normal.
In three separate experiments the O.N. response to 0.5M Na-Gluconate +
ASW applied to the osphradium was similar to that observed using IM Nacl- an
increased ammount of larger spikes. One example is shown in figure 11. On
the other hand, application of O.5M NMG-Cl + ASW yielded no definite effects
on O.N. activity. Figure 12 shows an example, and similar results were
obtained in two other experiments. The solution designed to produce a more
specific osmotic stress (0.6M Sucrose + ASW) evoked a response similar to
that caused by 0.5M Nacl + ASW, namely a large increase in the mid-sized
spikes (Fig 13). Although the reaction to O.54 Na-Gluconate t ASW is not as
profound as that in figure 11, there is still an increased number of the
larger sized spikes, but this effect appears to be shared by 0.6M Sucrose t
ASW.
One other solution proved to be a potent stimulus for the osphradium,
Im Licl elicited a response similar to that given by IM Nacl. There was a
great increase in amplitude and frequency following application of this
substance.
Substances which proved to be ineffective stimulants were light.
Pisaster ochraceous tube feet, dacron line (mechanical), and distilled water
(applied by the 250ul pipettor method).
DISCUSSION
The findings with the scanning electron microscope that the osphradium
of Colisella limatula does not show abundant sensory structures, in
particular, paddle cilia, appears to be congruent with data on the osphradium
of Patella and other prosobranch limpets. Possibly most sensory elements are
only free nerve endings buried under the microvillar coat, and very few
special sensory structures are visible protruding through this coat
(Haszprunar, 1985 A). However, the complete lack of ciliated sensory
elements noted in the present study must be questioned. This could possibly
be attributed to preparative techniques and would have to be verified with
additional observations of the osphradium under the scanning electron
microscope. Transmission electron microscopy would be necessary to observe
those areas beneath the microvilli-coated surface of the osphradium.
Electrical stimuli proved very useful in confirming that the recording
electrode made tight contact with the O.N.. Because the posterior edge and
12
medial corner of the osphradium were electrically sensitive it is likely that
the cell bodies found in these areas, using cobalt back fill, were
stimulating the O.N. to fire. These cell bodies possibly are constituents of
the osphradial ganglion.
Osphradial sensitivity to 1M Nacl, O.5M Nacl + ASW, O.SM Na-Gluconate.
and not to O.5M N-Methyl-Glucamine-Cl suggests that it is a chemosensor for
Na. However, because it is also sensitive to a high osmotic pressure (0.6M
Sucrose + ASW), with no increase in Na content, it also suggests that there
is a hyperosmotic receptor present in the osphradium. Why the organ does not
respond to hypertonic NMG-Cl is mysterious. Osphradial stimulation by
hypertonic media is a result similar to that observed in the opisthobranch
Aplysia (Stinnakre, J & L Tauc, 1968). The organ in Colisella does not
appear to respond to hypotonic sea water.
It is believed that the Na is not directly acting on the neural
processes near the osphradium increasing their excitability and causing the
increased O.N. activity. The cell bodies in question lie embedded in mantle
tissue, away from the surface of the organ. Excess flow from the solution
distributor does not directly reach the O.N. after it exits from the mantle
until long after the four seconds required for stimulation to start or
reverse. The flow is blocked by mantle tissue and the solution dissipates
away from the organ and nerve. Although additional controls for such
concerns are desirable, it is felt that the responses to chemical stimulation
are of fibers triggered by receptors in the organ proper.
Two effects of chemical stimulation are described in this report. The
first is an increase in both frequency and amplitude of the largest spikes
Obuv). Solutions evoking this response are 1M Nacl, O.5M Na-Gluconate +
ASW, and IM LiCl. The second effect is an increase in medium sized spikes
(3-6uv) due to the application of O.5M Nacl + ASW or 0.6M Sucrose + ASW.
I believe that one of two possibilities can account for these chemically
stimulated affects on the osphradium. The different effects on spike
amplitudes could be caused by different osphradial receptors stimulating
larger O.N. axons to fire in response to the 1M Nacl, O.5M'Na-Gluconate +
ASW, and 1M LiCl. This would result in larger action potentials than occur
normally. Smaller axons stimulated by other receptors for O.5M NaCl + ASW or
O.6M Sucrose + ASW would account for the effect of these solutions. Another
explanation for the two distinct responses involves a heightened frequency of
firing in one population of O.N. axons per unit sampling time. Even thouzh
the spike amplitude of single axons active during stimulation does not
increase, many additional axons may be activated thereby causing the
increased signal. Because data was displayed on a chart recorder with
limited frequency response it was not feasible to distinguish between these
two possibilities. Very fast, continuous sampling using either an
oscilloscope (recording on film) or a computerized data acquisition system
would resolve this uncertainty.
Further Cl, Na, and osmotic substitutes need to be tested to confirm the
results reported here and increase understanding of mechanisms underlying
these stimulants. Other intriguing substances such as food, gametes, body
wastes, and particulate matter beg investigation to uncover possible
influences on the osphradium, and thereby advance our knowledge of this
cryptic organ.
ACKNOWLEDGEMENTS
I would like to thank everyone associated with the
175H program for making this experience one of the best to
date. All of the professors which made this course possible
I extend an extra allotment of appreciation; Chuck Baxter.
Mark Dermy, and William Gilly. Thank you class for being with
me 25 hours a day. And to Tom Hahn our fearless T.A.
REFRENCES
Haszprunar, G. 1985. The fine morphology of the osphradial sense
organs of the mollusca. I. Gastropoda, Prosobranchia,
Philosophical Transactions of the Royal Society of London
307, 457-496.
Jahan-Parwar, B., M. Smith, R. VonBaumgarten. 1969. Activation
of neurosecretory cells in Aplysia by osphradial stimulation.
American Journal of Physiology 216, 1246-1256.
Stirnakre, J. and L. Tauc. 1969. Central neuronal response to
the activation of osmoreceptors in the osphradium of Aplysia.
Journal of Experimental Biology 51, 347-361.
lownsend, C.R. 1973. The role of the osphradium in chemoreception
by the snail Biomphalaria Glabrata (say). Animal Behavior 21,
549-556.
TABLE I.
Artificial Sea Water- 44OnM Nacl, 15nM Cacl,, 50nM NgCl,, 1OnM KCI.
10m Hepes, 967m osmols, PH-8,0.
O.5M Na-Gluconate + ASW-
1750m osmols, PH-8.1.
N-Methyl-Glucamine-Cl + ASW--
1850m osmols, PH-8 Titrated with 377 HCl
1M Naci--
1850m osmols.
0.6M Sucrose + ASW—
1742m osmols. PH-8.0.
1M LiC1-
1900m osmols.
Table I: Solutions used and their concentrations and osmolarities.
FIGURE LEGENDS
1. Colisella limatula, diagram of dissection; osphradium,
osphradial nerve, and location of recording electrode.
2. Osphradium as seen after CoCl, fill of the nerve trunk.
Area
where cell bodies are located'is most sensitive to electrical
stimulation.
3A. View of left osphradium under scanning electron microscope
magnified 850 times. Fissures are visible.
3B.
Surface of the left osphradium magnified 17k times. Surface
is composed of microvilli; cilia are not obvious, but possibly
present.
4A. Action potentials as seen due to electrical stimulation.
Voltages are between 7 and 10 volts for O.3ms.
4B. Spontaneous action potentials; these are of the same relative
size and time-course as the electrically stimulated ones,
5. Electrical stimulation of mantle adjacent to the osphradium
and corresponding lack of response.
Osphradial nerve activity due to the application of 250ul
of 1M NaCl.
7. Spike amplitude increase and decrease through time due to the
application of 1M NaCl through a hand micropipettor.
/(inset). Methods for measuring spike amplitude. Top half represents
method that assumes varying widths in the baseline. Bottom
half takes into account a constant baseline through time,
8A. Response of O.N. to the bathing of the osphradium with NSW.
8B. Response of O.N. to the changing of solutions from NSW to IM NaCl,
Larger slow spikes are caused by bubbles and movement of the organism.
80. Decrease in spike amplitude due to the addition of NSW.
9. Frequency of spikes greater than 2uv above the baseline in
response to the addition of NSW and 1M NaCl. Readings were
taken in 2 second blocks. Time to reach full response is
approximately 4 seconds after changing solutions.
10A. Number of spikes, occurring in four second blocks, vs. spike
amplitude before adjusting for constant baseline (see text
for details).
10B. Same as 10A after adjusting for a constant baseline.
11. Number of spikes vs. spike amplitude in response to the addition
of 1M NaCl, and O.5M Na-Gluconate + ASW.
12. Number of spikes as a function of amplitude due to the
application of O.5M NMG-CI +ASW, and O.5M NaCl + ASW.
13. Osphradial nerve response to the addition of O.6M Sucrose + ASW,
and 0.5M Na-Gluconate. Spike number vs. spike amplitude.
(
Osphradiug

Osph
Nerve




Recording Electrode
I


1
Scanning Electron Microscope
A
Ant
Osphradium
Ciliated Protozoa-









„Fissure
Microvilli
—Paddle Cilium?
—
C

—

I


—.—.—


-—
—
—

—

—



S
O
——




—-1
—


—






—


—





—









—
—.—

. —-


9
———
.

—


—.—
—
—-



—

—--


-.---.
:-


LECTRICAL






Mantle stimulation.



SOMV

—-—
—


—.—


*

—-—
—
MS



——
-—
i
——
—--- ----

—----- --
—.—





—


—
E

—-
—......
——
-—






——




1

















—----
--




i





——.—
—.-- —
--------.
O
L
-:
——-—

- - --
—
- â
L
10
—
..
L



o

-
+
——


—.—
5


0






0
—.———


—




(

-

â

. . -.











(






—


O










O








C


0
--—
—


O




.
Z




-

0





O
Z

1
—

1

















— —.—.—






S
—





1
+

S












—1—
















Z





—



OO

——



J
——
O







038 2 ud Salas
0
—
— -

—
-------
- - - -

A

NOT ADJUSTED
80
—



—

--
2 —




.

—
— ——40
—


.



-4








—1—

2
—




—





——
-





— —

9











——6


ME AMPLITUDE CIC


V.







—






+






R












ADJUSTED


—



.


â








—
—-- —

——






.—80
c0
L
O

——

- -.-
..
—

---------—
——
40
S

1



A

------.---—.—
/



-



4

. . ..


6

SPIKE AMPLITUDE CIO V.)



NSW
. .*
—
—---
5M Naci & ASW
*-- --.—--
——--- -
—.

MNaci

--- -
—-———
—
t
— —
—--
——-
O


O
L
0 u



--
—


—
—.—


























(O





























—

——
— ———





























































—

—











—







8
—
SNIdS JO ON
---.
—-—
——

—



Z







C



——






























-

















—











—









1






—:—

















—


:


—

SANIdS JO ON



—
—


—

(




















1
——.—
e



—- —--



20
—----
C
0
—--——






9

—-
0
—
——

—



9







——

â












—







—















---




—
S
—





10













—

—


—






4





















8

SANIdS JO ON
—