Visit https://testbankdeal.

com to download the full version and

explore more testbank or solution manual

Data Structures And Algorithm Analysis In C++

4th Edition Weiss Solutions Manual

_____ Click the link below to download _____

https://testbankdeal.com/product/data-structures-and-
algorithm-analysis-in-c-4th-edition-weiss-solutions-
manual/

Explore and download more testbank at testbankdeal.com

Here are some suggested products you might be interested in.
Click the link to download

Data Structures And Algorithm Analysis In Java 3rd Edition

Weiss Solutions Manual

https://testbankdeal.com/product/data-structures-and-algorithm-
analysis-in-java-3rd-edition-weiss-solutions-manual/

Data Structures and Algorithms in C++ 2nd Edition Goodrich

Solutions Manual

https://testbankdeal.com/product/data-structures-and-algorithms-
in-c-2nd-edition-goodrich-solutions-manual/

Introduction to C++ Programming and Data Structures 4th

Edition Liang Solutions Manual

https://testbankdeal.com/product/introduction-to-c-programming-and-
data-structures-4th-edition-liang-solutions-manual/

Pharmacology for Nurses A Pathophysiologic Approach 5th

Edition Adams Test Bank

https://testbankdeal.com/product/pharmacology-for-nurses-a-
pathophysiologic-approach-5th-edition-adams-test-bank/
Business Communication Developing Leaders for a Networked
World 1st Edition Cardon Solutions Manual

https://testbankdeal.com/product/business-communication-developing-
leaders-for-a-networked-world-1st-edition-cardon-solutions-manual/

Financial Accounting The Impact on Decision Makers 8th

Edition Porter Solutions Manual

https://testbankdeal.com/product/financial-accounting-the-impact-on-
decision-makers-8th-edition-porter-solutions-manual/

Byrd and Chens Canadian Tax Principles 2018-2019 1st

Edition Byrd Test Bank

https://testbankdeal.com/product/byrd-and-chens-canadian-tax-
principles-2018-2019-1st-edition-byrd-test-bank/

Contemporary Accounting A Strategic Approach for Users 9th

Edition Hancock Solutions Manual

https://testbankdeal.com/product/contemporary-accounting-a-strategic-
approach-for-users-9th-edition-hancock-solutions-manual/

Microeconomics 2nd Edition Bernheim Solutions Manual

https://testbankdeal.com/product/microeconomics-2nd-edition-bernheim-
solutions-manual/
Molecular Biology of The Cell 5th edition Alberts Test
Bank

https://testbankdeal.com/product/molecular-biology-of-the-cell-5th-
edition-alberts-test-bank/
CHAPTER 6

Priority Queues (Heaps)

6.1 Yes. When an element is inserted, we compare it to the current minimum and change the minimum if the new

element is smaller. deleteMin operations are expensive in this scheme.

6.2

6.3 The result of three deleteMins, starting with both of the heaps in Exercise 6.2, is as follows:

6.4 (a) 4N

(b) O(N2)

(c) O(N4.1)

(d) O(2N)

6.5
/**
* Insert item x, allowing duplicates.
*/
 void insert( const Comparable & x )
{
if( currentSize == array.size( ) - 1 )
array.resize( array.size( ) * 2 );
// Percolate up
int hole = ++currentSize;
for( ; hole > 1 && x < array[ hole / 2 ]; hole /= 2 )
array[ hole ] = array[ hole / 2 ];
array[0] = array[ hole ] = x;
}

6.6 225. To see this, start with i = 1 and position at the root. Follow the path toward the last node, doubling i

when taking a left child, and doubling i and adding one when taking a right child.

6.7 (a) We show that H(N), which is the sum of the heights of nodes in a complete binary tree of N nodes, is

N − b(N), where b(N) is the number of ones in the binary representation of N. Observe that for N = 0 and

N = 1, the claim is true. Assume that it is true for values of k up to and including N − 1. Suppose the left and

right subtrees have L and R nodes, respectively. Since the root has height  log N  , we have

H (N ) = log N  + H (L) + H (R )
= log N  + L − b(L) + R − b(R)
= N − 1 + (  Log N  − b(L) − b(R) )

The second line follows from the inductive hypothesis, and the third follows because L + R = N − 1. Now the

last node in the tree is in either the left subtree or the right subtree. If it is in the left subtree, then the right

subtree is a perfect tree, and b(R) = log N  − 1 . Further, the binary representation of N and L are identical,

with the exception that the leading 10 in N becomes 1 in L. (For instance, if N = 37 = 100101, L = 10101.) It

is clear that the second digit of N must be zero if the last node is in the left subtree. Thus in this case,

b(L) = b(N), and

H(N) = N − b(N)

If the last node is in the right subtree, then b(L) =  log N  . The binary representation of R is identical to

N, except that the leading 1 is not present. (For instance, if N = 27 = 101011, L = 01011.) Thus

b(R) = b(N) − 1, and again

H(N) = N − b(N)
 (b) Run a single-elimination tournament among eight elements. This requires seven comparisons and

generates ordering information indicated by the binomial tree shown here.

The eighth comparison is between b and c. If c is less than b, then b is made a child of c. Otherwise, both

c and d are made children of b.

(c) A recursive strategy is used. Assume that N = 2k. A binomial tree is built for the N elements as in part (b).

The largest subtree of the root is then recursively converted into a binary heap of 2 k − 1 elements. The last

element in the heap (which is the only one on an extra level) is then inserted into the binomial queue

consisting of the remaining binomial trees, thus forming another binomial tree of 2 k − 1 elements. At that

point, the root has a subtree that is a heap of 2 k − 1 − 1 elements and another subtree that is a binomial tree of

2k−1 elements. Recursively convert that subtree into a heap; now the whole structure is a binary heap. The

running time for N = 2k satisfies T(N) = 2T(N/2) + log N. The base case is T(8) = 8.

6.8 a) Since each element in a min heap has children whose elements are greater than the value in the
element itself, the maximum element has no children and is a leaf.

c) Since the maximum element can be any leaf (the position of a node is determined entirely by the
value of its parent and children), all leaves must be examined to find the maximum value in a min
heap.

6.9 Let D1, D2, . . . ,Dk be random variables representing the depth of the smallest, second smallest, and kth

smallest elements, respectively. We are interested in calculating E(Dk). In what follows, we assume that the

heap size N is one less than a power of two (that is, the bottom level is completely filled) but sufficiently

large so that terms bounded by O(1/N) are negligible. Without loss of generality, we may assume that the kth

smallest element is in the left subheap of the root. Let pj, k be the probability that this element is the jth

smallest element in the subheap.

Lemma.
 k −1
For k > 1, E (Dk ) =  p j ,k (E (D j ) + 1) .
j =1

Proof.

An element that is at depth d in the left subheap is at depth d + 1 in the entire subheap. Since

E(Dj + 1) = E(Dj) + 1, the theorem follows.

Since by assumption, the bottom level of the heap is full, each of second, third, . . . , k − 1th smallest

elements are in the left subheap with probability of 0.5. (Technically, the probability should be half − 1/(N −

1) of being in the right subheap and half + 1/(N − 1) of being in the left, since we have already placed the kth

smallest in the right. Recall that we have assumed that terms of size O(1/N) can be ignored.) Thus

1  k − 2
p j ,k = pk − j ,k = k −2  
2  j −1 

Theorem.

E(Dk)  log k.

Proof.

The proof is by induction. The theorem clearly holds for k = 1 and k = 2. We then show that it holds for

arbitrary k > 2 on the assumption that it holds for all smaller k. Now, by the inductive hypothesis, for any

1  j  k − 1,

E (D j ) + E (Dk − j )  log j + log k − j

Since f(x) = log x is convex for x > 0,

log j + log k − j  2 log ( k 2 )

Thus

E (D j ) + E (Dk − j )  log( k 2 ) + log ( k 2 )

Furthermore, since pj, k = pk − j, k,

p j ,k E (D j ) + pk − j ,k E (Dk − j )  p j ,k log ( k 2 ) + pk − j ,k log ( k 2 )

From the lemma,

 k −1
E (Dk ) =  p j , k (E (D j ) + 1)
j =1
k −1
=1+  p j, k E(D j )
j =1

Thus

k −1
E ( Dk )  1 +  p j , k log ( k 2)
j =1
k −1
 1 + log ( k 2)  p j , k
j =1

 1 + log ( k 2)
 log k

completing the proof.

It can also be shown that asymptotically, E(Dk)  log(k − 1) − 0.273548.

6.10 (a) Perform a preorder traversal of the heap.

(b) Works for leftist and skew heaps. The running time is O(Kd) for d-heaps.

6.12 Simulations show that the linear time algorithm is the faster, not only on worst-case inputs, but also on

random data.

6.13 (a) If the heap is organized as a (min) heap, then starting at the hole at the root, find a path down to a leaf by

taking the minimum child. The requires roughly log N comparisons. To find the correct place where to move

the hole, perform a binary search on the log N elements. This takes O(log log N) comparisons.

(b) Find a path of minimum children, stopping after log N − log log N levels. At this point, it is easy to

determine if the hole should be placed above or below the stopping point. If it goes below, then continue

finding the path, but perform the binary search on only the last log log N elements on the path, for a total of

log N + log log log N comparisons. Otherwise, perform a binary search on the first log N − log log N

elements. The binary search takes at most log log N comparisons, and the path finding took only log N − log

log N, so the total in this case is log N. So the worst case is the first case.

(c) The bound can be improved to log N + log*N + O(1), where log*N is the inverse Ackerman function (see

Chapter 8). This bound can be found in reference [17].

6.14 The parent is at position (i + d − 2) d  . The children are in positions (i − 1)d + 2, . . . , id + 1.

6.15 (a) O((M + d N) logd N).

 (b) O((M + N) log N).

(c) O(M + N2).

(d) d = max(2, M/N). (See the related discussion at the end of Section 11.4.)

6.16 Starting from the second most signficant digit in i, and going toward the least significant digit, branch left for

0s, and right for 1s.

6.17 (a) Place negative infinity as a root with the two heaps as subtrees. Then do a deleteMin.

(b) Place negative infinity as a root with the larger heap as the left subheap, and the smaller heap as the right

subheap. Then do a deleteMin.

(c) SKETCH: Split the larger subheap into smaller heaps as follows: on the left-most path, remove two

subheaps of height r − 1, then one of height r, r + 1, and so one, until l − 2. Then merge the trees, going

smaller to higher, using the results of parts (a) and (b), with the extra nodes on the left path substituting for

the insertion of infinity, and subsequent deleteMin.

6.18 a. The minimum element will be the root. The maximum element will be one of the two children of the root.

b. Place the new element in the last open position (as in a regular heap). Now compare it to its parent. Now if
the new element was inserted into a max(min) row and was less (greater) than its parent. Then swap it with
its parent and now it need only be compared to other min elements and bubbled up min elements. If it is
greater (less) than its parent it need only be compared to other max elements bubbled up using the max
elements.

c) For a min deletion, remove the root . Let the last element in the heap be x . If the root has no children, then
x becomes the root. If find m the minimum child or grandchild of the root. If k Y m, then x becomes the
root. Other wise if m is the child of the root the m becomes the root and x is inserted in place of m. Finally if
m is the grandchild of the root, then m is moved to the root and if p is the parent of m, then if x > is p, then p
and x are interchanged.

d) yes. (see Atkinson et al., Min-Max Heaps and Generalized Priority Queues, Programming Techniques
and Data Structures, Vol 29, No. 10, pp. 996 - 1000, 1986.)
6.20

6.21 This theorem is true, and the proof is very much along the same lines as Exercise 4.20.

6.22 If elements are inserted in decreasing order, a leftist heap consisting of a chain of left children is formed. This

is the best because the right path length is minimized.

6.23 (a) If a decreaseKey is performed on a node that is very deep (very left), the time to percolate up would be

prohibitive. Thus the obvious solution doesn’t work. However, we can still do the operation efficiently by a

combination of remove and insert. To remove an arbitrary node x in the heap, replace x by the merge of its

left and right subheaps. This might create an imbalance for nodes on the path from x’s parent to the root that

would need to be fixed by a child swap. However, it is easy to show that at most logN nodes can be affected,

preserving the time bound.

This is discussed in Chapter 11.

6.24 Lazy deletion in leftist heaps is discussed in the paper by Cheriton and Tarjan [10]. The general idea is that if

the root is marked deleted, then a preorder traversal of the heap is formed, and the frontier of marked nodes is

removed, leaving a collection of heaps. These can be merged two at a time by placing all the heaps on a

queue, removing two, merging them, and placing the result at the end of the queue, terminating when only

one heap remains.

6.25 (a) The standard way to do this is to divide the work into passes. A new pass begins when the first element

reappears in a heap that is dequeued. The first pass takes roughly 2*1*(N/2) time units because there are N/2

merges of trees with one node each on the right path. The next pass takes 2*2*(N/4) time units because of the

roughly N/4 merges of trees with no more than two nodes on the right path. The third pass takes 2*3*(N/8)

time units, and so on. The sum converges to 4N.

(b) It generates heaps that are more leftist.

6.26

6.27

6.28 This claim is also true, and the proof is similar in spirit to Exercise 4.20 or 6.21.

6.29 Yes. All the single operation estimates in Exercise 6.25 become amortized instead of worst-case, but by the

definition of amortized analysis, the sum of these estimates is a worst-case bound for the sequence.

6.30 Clearly the claim is true for k = 1. Suppose it is true for all values i = 1, 2, . . . , k. A Bk + 1 tree is formed by

attaching a Bk tree to the root of a Bk tree. Thus by induction, it contains a B0 through Bk − 1 tree, as well as the

newly attached Bk tree, proving the claim.

6.31 Proof is by induction. Clearly the claim is true for k = 1. Assume true for all values i = 1, 2, . . . ,k. A Bk + 1

k 
tree is formed by attaching a Bk tree to the original Bk tree. The original thus had   nodes at depth d. The
d 

 k 
attached tree had   nodes at depth d−1, which are now at depth d. Adding these two terms and using a
 d − 1

well-known formula establishes the theorem.

6.32

6.33 This is established in Chapter 11.

6.34

template<typename Comparable>
struct BiQueNode
{
Comparable item;
vector<BiQueNode *> pointers;
BiQueNode<Comparable> (Comparable e) : item(e) {}
};
template <typename Comparable>
class BinomalQue
{
private:
vector<BiQueNode<Comparable>> biQue;
};

template <typename Comparable>

BiQueNode<Comparable> * combine(BiQueNode<Comparable> * p, BiQueNode<Comparable> * q)
{
if (p->item < q->item)
{
p->pointers.push_back(q);
return p;
}
else
{
q->pointers.push_back(p);
return q;
}
}

template <typename Comparable>

BiQueNode<Comparable> * insert(Comparable v)
{
 BiQueNode<Comparable> * t = new BiQueNode<Comparable> v;
BiQueNode<Comparable> * c = t;
for (int i = 0; i <= biQue.size(); i++)
{
if (c == nullptr) break;
if (i == biQue.size() -1)
biQue.push_back(nullptr);
if (biQue[i] == nullptr)
{ biQue[i] = c; break;}
c = combine(c, bq[i]);
bique[i] = null;
}
return t;
}

6.37

/*
Bin packing
*/
#include <vector>
#include <queue>
using namespace std;
const double Cap = 1.0;
class Bins
{
private:
vector<double> bins;
priority_queue<double> heapBins;
public:
Bins(int size = 0)
{bins.resize(size);
for (int i = 0; i < size; i++)
bins[i] = 0;
}
void clear() {bins.clear();}
int size(){ return bins.size();}
void insertFirstFit(double item) // a.
{
for (int i = 0; i < bins.size(); i++)
if (bins[i] + item < Cap)
{
bins[i] += item;
return;
}
bins.push_back(item);
}
int insertWorstFit(double item) // b
{
static int size = 0;
double maxRoom;
if (heapBins.empty())
heapBins.push(Cap - item);
else
{
maxRoom = heapBins.top();
if (maxRoom > item) // there is room
{
Visit https://testbankdead.com
now to explore a rich
collection of testbank,
solution manual and enjoy
exciting offers!
 heapBins.pop();
heapBins.push(maxRoom - item);
}
else
{heapBins.push(Cap - item);
size++;
}
}
return size;
}
void insertBestFit(double item) // c
{
double gap =Cap;
int gapIndex = -1;
if (bins.size() == 0)
bins.push_back(item);
else
{
for (int i = 0; i < bins.size(); i++)
if (bins[i]+item < Cap && bins[i]+item < gap)
{
gap = Cap - bins[i] - item;
gapIndex = i;
}
if (gapIndex < 0)
bins.push_back(item);
else
bins[gapIndex] += item;
}
}
};

d. yes

6.38 Don’t keep the key values in the heap, but keep only the difference between the value of the key in a node

and the value of the parent’s key.

6.39 O(N + k log N) is a better bound than O(N log k). The first bound is O(N) if k = O(N/log N). The second

bound is more than this as soon as k grows faster than a constant. For the other values (N/log N) = k = (N),

the first bound is better. When k = (N), the bounds are identical.
Exploring the Variety of Random
Documents with Different Content
shown at o f. When the knife-edges are suitably placed the frame is
balanced, so that a small piece of paper laid at a will cause that side
to descend.

Fig. 24.
77. We suspend two small hooks from the points a and b: these
are made of fine wire, so that their weight may be left out of
consideration. With this apparatus we can in the first place verify the
principle of equality of moments: for example, if I place the hook a
at a distance of 9" from the centre o and load it with 1 lb., I find that
when b is laden with 0·5 lb. it must be at a distance of 18" from o in
order to counterbalance a; the moment in the one case is 9 × 1, in
the other 18 × 0·5, and these are obviously equal.
78. Let a weight of 1 lb. be placed on each of the hooks, the
frame will only be in equilibrium when the hooks are at precisely the
same distance from the centre. A familiar application of this principle
is found in the ordinary weighing scales; the frame, which in this
case is called a beam, is sustained by two knife-edges, smaller,
however, than those represented in the figure. The pans p, p are
suspended from the extremities of the beam, and should be at equal
distances from its centre. These scale-pans must be of equal weight,
and then, when equal weights are placed in them, the beam will
remain horizontal. If the weight in one slightly exceed that in the
other, the pan containing the heavier weight will of course descend.
79. That a pair of scales should weigh accurately, it is necessary
that the weights be correct; but even with correct weights, a balance
of defective construction will give an inaccurate result. The error
frequently arises from some inequality in the lengths of the arms of
the beam. When this is the case, the two weights which really
balance are not equal. Supposing, for instance, that with an
imperfect balance I endeavour to weigh a pound of shot. If I put the
weight on the short side, then the quantity of shot balanced is less
than 1 lb.; while if the 1 lb. weight be placed at the long side, it will
require more than 1 lb. of shot to produce equilibrium. The mode of
testing a pair of scales is then evident. Let weights be placed in the
pans which balance each other; if the weights be interchanged and
the balance still remains horizontal, it is correct.
80. Suppose, for example, that the two arms be 10 inches and 11
inches long, then, if 1 lb. weight be placed in the pan of the 10-inch
end, its moment is 10; and if ¹⁰/₁₁ of 1 lb. be placed in the pan
belonging to the 11-inch end, its moment is also 10: hence 1 lb. at
the short end balances ¹⁰/₁₁ of 1 lb. at the long end; and therefore,
if the shopkeeper placed his weight in the short arm, his customers
would lose ¹/₁₁ part of each pound for which they paid; on the other
hand, if the shopkeeper placed his 1 lb. weight on the long arm,
then not less than ¹¹/₁₀ lb. would be required in the pan belonging
to the short arm. Hence in this case the customer would get ¹/₁₀ lb.
too much. It follows, that if a shopman placed his weights and his
goods alternately in the one scale and in the other he would be a
loser on the whole; for, though every second customer gets ¹/₁₁ lb.
less than he ought, yet the others get ¹/₁₀ lb. more than they have
paid for.
 LECTURE IV.
THE FORCE OF GRAVITY.

Introduction.—Specific Gravity.—The Plummet

and Spirit-Level.—The Centre of Gravity.—
Stable and Unstable Equilibrium.—Property
of the Centre of Gravity in a Revolving
Wheel.

INTRODUCTION.
81. In the last three lectures we considered forces in the
abstract; we saw how they are to be represented by straight lines,
how compounded together and how decomposed into others; we
have explained what is meant by forces being in equilibrium, and we
have shown instances where the forces lie in the same plane or in
different planes, and where they intersect or are parallel to each
other. These subjects are the elements of mechanics; they form the
framework which in this and subsequent lectures we shall try to
present in a more attractive garb. We shall commence by studying
the most remarkable force in nature, a force constantly in action,
and one to which all bodies are subject, a force which distance
cannot annihilate, and one the properties of which have led to the
most sublime discoveries of human intellect. This is the force of
gravity.
82. If I drop a stone from my hand, it falls to the ground. That
which produces motion is a force: hence the stone must have been
acted upon by a force which drew it to the ground. On every part of
the earth’s surface experience shows that a body tends to fall. This
fact proves that there is an attractive force in the earth tending to
draw all bodies towards it.

Fig. 25.
83. Let a b c d (Fig. 25) be points from which stones are let fall,
and let the circle represent the section of the earth; let p q r s be the
points at the surface of the earth upon which the stones will drop
when allowed to do so. The four stones will move in the directions of
the arrows: from a to p the stone moves in an opposite direction to
the motion from c to r; from b to q it moves from right to left, while
from d to s it moves from left to right. The movements are in
different directions; but if I produce these directions, as indicated by
the dotted lines, they each pass through the centre o.
84. Hence each stone in falling moves towards the centre of the
earth, and this is consequently the direction of the force. We
therefore assert that the earth has an attraction for the stone, in
consequence of which it tries to get as near the earth’s centre as
possible, and this attraction is called the force of gravitation.
 85. We are so excessively familiar with the phenomenon of
seeing bodies fall that it does not excite our astonishment or arouse
our curiosity. A clap of thunder, which every one notices, because
much less frequent, is not really more remarkable. We often look
with attention at the attraction of a piece of iron by a magnet, and
justly so, for the phenomenon is very interesting, and yet the falling
of a stone is produced by a far grander and more important force
than the force of magnetism.
86. It is gravity which causes the weight of bodies. I hold a
piece of lead in my hand: gravity tends to pull it downwards, thus
producing a pressure on my hand which I call weight. Gravity acts
with slightly varying intensity at various parts of the earth’s surface.
This is due to two distinct causes, one of which may be mentioned
here, while the other will be subsequently referred to. The earth is
not perfectly spherical; it is flattened a little at the poles;
consequently a body at the pole is nearer the general mass of the
earth than a body at the equator; therefore the body at the pole is
more attracted, and seems heavier. A mass which weighs 200 lbs. at
the equator would weigh one pound more at the pole: about one-
third of this increase is due to the cause here pointed out. (See
Lecture XVII.)
87. Gravity is a force which attracts every particle of matter; it
acts not merely on those parts of a body which lie on the surface,
but it equally affects those in the interior. This is proved by observing
that a body has the same weight, however its shape be altered: for
example, suppose I take a ball of putty which weighs 1 lb., I shall
find that its weight remains unchanged when the ball is flattened
into a thin plate, though in the latter case the surface, and therefore
the number of superficial particles, is larger than it was in the
former.

SPECIFIC GRAVITY.
 88. Gravity produces different effects upon different substances.
This is commonly expressed by saying that some substances are
heavier than others; for example, I have here a piece of wood and a
piece of lead of equal bulk. The lead is drawn to the earth with a
greater force than the wood. Substances are usually termed heavy
when they sink in water, and light when they float upon it. But a
body sinks in water if it weighs more than an equal bulk of water,
and floats if it weigh less. Hence it is natural to take water as a
standard with which the weights of other substances may be
compared.
89. I take a certain volume, say a cubic inch of cast iron such as
this I hold in my hand, and which has been accurately shaped for
the purpose. This cube is heavier than one cubic inch of water, but I
shall find that a certain quantity of water is equal to it in weight;
that is to say, a certain number of cubic inches of water, and it may
be fractional parts of a cubic inch, are precisely of the same weight.
This number is called the specific gravity of cast iron.
90. It would be impossible to counterpoise water with the iron
without holding the water in a vessel, and the weight of the vessel
must then be allowed for. I adopt the following plan. I have here a
number of inch cubes of wood (Fig. 26), which would each be lighter
than a cubic inch of water, but I have weighted the wooden cubes
by placing grains of shot into holes bored into the wood. The weight
of each cube has thus been accurately adjusted to be equal to that
of a cubic inch of water. This may be tested by actual weighing. I
weigh one of the cubes and find it to be 252 grains, which is well
known to be the weight of a cubic inch of water.
 Fig. 26.
91. But the cubes may be shown to be identical in weight with
the same bulk of water by a simpler method. One of them placed in
water should have no tendency to sink, since it is not heavier than
water, nor on the other hand, since it is not lighter, should it have
any tendency to float. It should then remain in the water in
whatever position it may be placed. It is difficult to prepare one of
these cubes so accurately that this result should be attained, and it
is impossible to ensure its continuance for any time owing to
changes of temperature and the absorption of water by the wood.
We can, however, by a slight modification, prove that one of these
cubes is at all events nearly equal in weight to the same bulk of
water. In Fig. 26 is shown a tall glass jar filled with a fluid in
appearance like plain water, but it is really composed in the following
manner. I first poured into the jar a very weak solution of salt and
water, which partially filled it; I then poured gently upon this a little
pure water, and finally filled up the jar with water containing a little
spirits of wine: the salt and water is a little heavier than pure water,
while the spirit and water is a little lighter. I take one of the cubes
and drop it gently into the glass; it falls through the spirit and water,
and after making a few oscillations settles itself at rest in the
stratum shown in the figure. This shows that our prepared cube is a
little heavier than spirit and water, and a little lighter than salt and
water, and hence we infer that it must at all events be very near the
weight of pure water which lies between the two. We have also a
number of half cubes, quarter cubes, and half-quarter cubes, which
have been similarly prepared to be of equal weight with an equal
bulk of water.
92. We shall now be able to measure the specific gravity of a
substance. In one pan of the scales I place the inch cube of cast
iron, and I find that 7¼ of the wooden cubes, which we may call
cubes of water, will balance it. We therefore say that the specific
gravity of iron is 7¼. The exact number found by more accurate
methods is 7·2. It is often convenient to remember that 23 cubic
inches of cast iron weigh 6 lbs., and that therefore one cubic inch
weighs very nearly ¼ lb.
93. I have also cubes of brass, lead, and ivory; by counterpoising
them with the cubes of water, we can easily find their specific
gravities; they are shown together with that of cast iron in the
following table:—
Substance. Specific Gravity.
Cast Iron 7·2
Brass 8·1
Lead 11·3
Ivory 1·8
94. The mode here adopted of finding specific gravities is entirely
different from the far more accurate methods which are commonly
used, but the explanation of the latter involve more difficult
principles than those we have been considering. Our method rather
offers an explanation of the nature of specific gravity than a good
means of determining it, though, as we have seen, it gives a result
sufficiently near the truth for many purposes.

THE PLUMMET AND SPIRIT-LEVEL.

95. The tendency of the earth to draw all bodies towards it is
well illustrated by the useful “line and plummet.” This consists
merely of a string to one end of which a leaden weight is attached.
The string when at rest hangs vertically; if the weight be drawn to
one side, it will, when released, swing backwards and forwards, until
it finally settles again in the vertical; the reason is that the weight
always tries to get as near the earth as it can, and this is
accomplished when the string hangs vertically downwards.
96. The surface of water in equilibrium is a horizontal plane; that
is also a consequence of gravity. All the particles of water try to get
as near the earth as possible, and therefore if any portion of the
water were higher than the rest, it would immediately spread, as by
doing so it could get lower.
97. Hence the surface of a fluid at rest enables us to find a
perfectly horizontal plane, while the plummet gives us a perfectly
vertical line: both these consequences of gravity are of the utmost
practical importance.
98. The spirit-level is another common and very useful
instrument which depends on gravity. It consists of a glass tube
slightly curved, with its convex surface upwards, and attached to a
stand with a flat base. This tube is nearly filled with spirit, but a
bubble of air is allowed to remain. The tube is permanently adjusted
so that, when the plate is laid on a perfectly horizontal surface, the
bubble will stand in the middle: accordingly the position of the
bubble gives a means of ascertaining whether a surface is level.

THE CENTRE OF GRAVITY.

 99. We proceed to an experiment which
will give an insight into a curious property of
gravity. I have here a plate of sheet iron; it
has the irregular shape shown in Fig. 27.
Five small holes a b c d e are punched at
different positions on the margin. Attached
to the framework is a small pin from which I
can suspend the iron plate by one of its
holes a: the plate is not supported in any
other way; it hangs freely from the pin,
around which it can be easily turned. I find
that there is one position, and one only, in
which the plate will rest; if I withdraw it from
that position it returns there after a few
oscillations. In order to mark this position, I
suspend a line and plummet from the pin,
Fig. 27.
having rubbed the line with chalk. I allow the
line to come to rest in front of the plate. I then flip the string against
the plate, and thus produce a chalked mark: this of course traces
out a vertical line a p on the plate.
I now remove the plummet and suspend the plate from another
of its holes b, and repeat the process, thus drawing a second chalked
line b p across the plate, and so on with the other holes: I thus
obtain five lines across the plate, represented by dotted lines in the
figure. It is a very remarkable circumstance that these five lines all
intersect in the same point p; and if additional holes were bored in
the plate, whether in the margin or not, and the chalk line drawn
from each of them in the manner described, they would one and all
pass through the same point. This remarkable point is called the
centre of gravity of the plate, and the result at which we have
arrived may be expressed by saying that the vertical line from the
point of suspension always passes through the centre of gravity.
100. At the centre of gravity p a hole has been bored, and when I
place the supporting pin through this hole you see that the plate will
rest indifferently in all positions: this is a curious property of the
centre of gravity. The centre of gravity may in this respect be
contrasted with another hole q, which is only an inch distant: when I
support the plate by this hole, it has only one position of rest, viz.
when the centre of gravity p is vertically beneath q. Thus the centre
of gravity differs remarkably from any other point in the plate.
101. We may conceive the force of gravity on the plate to act as
a force applied at p. It will then be easily seen why this point
remains vertically underneath the point of suspension when the
body is at rest. If I attached a string to the plate and pulled it, the
plate would evidently place itself so that the direction of the string
would pass through the point of suspension; in like manner gravity
so places the plate that the direction of its force passes through the
point of suspension.
102. Whatever be the form of the plate it always contains one
point possessing these remarkable properties, and we may state in
general that in every body, no matter what be its shape, there is a
point called the centre of gravity, such that if the body be suspended
from this point it will remain in equilibrium indifferently in any
position, and that if the body be suspended from any other point,
then it will be in equilibrium when the centre of gravity is directly
underneath the point of suspension. In general, it will be impossible
to support a body exactly at its centre of gravity, as this point is
within the mass of the body, and it may also sometimes happen that
the centre of gravity does not lie in the substance of the body at all,
as for example in a ring, in which case the centre of gravity is at the
centre of the ring. We need not, however, dwell on these exceptional
cases, as sufficient illustrations of the truth of the laws mentioned
will present themselves subsequently.

STABLE AND UNSTABLE EQUILIBRIUM.

103. An iron rod a b, capable of revolving round an axis passing
through its centre p, is shown in Fig. 28.
 The centre of gravity lies at the centre b, and
consequently, as is easily seen, the rod will remain at rest
in whatever position it be placed. But let a weight r be
attached to the rod by means of a binding screw. The
centre of gravity of the whole is no longer at the centre of
the rod; it has moved to a point s nearer the weight; we
may easily ascertain its position by removing the rod from
its axle and then ascertaining the point about which it will
balance. This may be done by placing the bar on a knife-
edge, and moving it to and fro until the right position be
secured; mark this position on the rod, and return it to its
axle, the weight being still attached. We do not now find
that the rod will balance in every position. You see it will
balance if the point s be directly underneath the axis, but
not if it lie to one side or the other. But if s be directly over
the axis, as in the figure, the rod is in a curious condition.
It will, when carefully placed, remain at rest; but if it
receive the slightest displacement, it will tumble over. The
rod is in equilibrium in this position, but it is what is called
unstable equilibrium. If the centre of gravity be vertically
Fig. 28. below the point of suspension, the rod will return again if
moved away: this position is therefore called one of stable
equilibrium. It is very important to notice the distinction between
these two kinds of equilibrium.
104. Another way of stating the case is as follows. A body is in
stable equilibrium when its centre of gravity is at the lowest point:
unstable when it is at the highest. This may be very simply
illustrated by an ellipse, which I hold in my hand. The centre of
gravity of this figure is at its centre. The ellipse, when resting on its
side, is in a position of stable equilibrium and its centre of gravity is
then clearly at its lowest point. But I can also balance the ellipse on
its narrow end, though if I do so the smallest touch suffices to
overturn it. The ellipse is then in unstable equilibrium; in this case,
obviously, the centre of gravity is at the highest point.
 105. I have here a sphere, the centre of gravity
of which is at its centre; in whatever way the
sphere is placed on a plane, its centre is at the
same height, and therefore cannot be said to have
any highest or lowest point; in such a case as this
the equilibrium is neutral. If the body be
displaced, it will not return to its old position, as it
would have done had that been a position of
stable equilibrium, nor will it deviate further
therefrom as if the equilibrium had been unstable: Fig. 29.
it will simply remain in the new position to which it
is brought.
106. I try to balance an iron ring upon the end of a stick h, Fig.
29, but I cannot easily succeed in doing so. This is because its
centre of gravity s is above the point of support; but if I place the
stick at f, the ring is in stable equilibrium, for now the centre of
gravity is below the point of support.

PROPERTY OF THE CENTRE OF GRAVITY

IN A REVOLVING WHEEL.
107. There are other curious consequences which follow from the
properties of the centre of gravity, and we shall conclude by
illustrating one of the most remarkable, which is at the same time of
the utmost importance in machinery.
 Fig. 30.
108. It is generally necessary that a machine should work as
steadily as possible, and that undue vibration and shaking of the
framework should be avoided: this is particularly the case when any
parts of the machine rotate with great velocity, as, if these be heavy,
inconvenient vibration will be produced when the proper
adjustments are not made. The connection between this and the
centre of gravity will be understood by reference to the apparatus
represented in the accompanying figure (Fig. 30). We have here an
arrangement consisting of a large cog wheel c working into a small
one b, whereby, when the handle h is turned, a velocity of rotation
can be given to the iron disk d, which weighs 14 lbs, and is 18" in
diameter. This disk being uniform, and being attached to the axis at
its centre, it follows that its centre of gravity is also the centre of
rotation. The wheels are attached to a stand, which, though
massive, is still unconnected with the floor. By turning the handle I
can rotate the disk very rapidly, even as much as twelve times in a
second. Still the stand remains quite steady, and even the shutter
bell attached to it at e is silent.
109. Through one of the holes in the disk d I fasten a small iron
bolt and a few washers, altogether weighing about 1 lb.; that is, only
one-fourteenth of the weight of the disk. When I turn the handle
slowly, the machine works as smoothly as before; but as I increase
the speed up to one revolution every two seconds, the bell begins to
ring violently, and when I increase it still more, the stand quite
shakes about on the floor. What is the reason of this? By adding the
bolt, I slightly altered the position of the centre of gravity of the
disk, but I made no change of the axis about which the disk rotated,
and consequently the disk was not on this occasion turning round its
centre of gravity: this it was which caused the vibration. It is
absolutely necessary that the centre of gravity of any heavy piece,
rotating rapidly about an axis, should lie in the axis of rotation. The
amount of vibration produced by a high velocity may be very
considerable, even when a very small mass is the originating cause.
110. In order that the machine may work smoothly again, it is
not necessary to remove the bolt from the hole. If by any means I
bring back the centre of gravity to the axis, the same end will be
attained. This is very simply effected by placing a second bolt of the
same size at the opposite side of the disk, the two being at equal
distances from the axis; on turning the handle, the machine is seen
to work as smoothly as it did in the first instance.
111. The most common rotating pieces in machines are wheels
of various kinds, and in these the centre of gravity is evidently
identical with the centre of rotation; but if from any cause a wheel,
which is to turn rapidly, has an extra weight attached to one part,
this weight must be counterpoised by one or more on other portions
of the wheel, in order to keep the centre of gravity of the whole in
its proper place. Thus it is that the driving wheels of a locomotive
are always weighted so as to counteract the effect of the crank and
restore the centre of gravity to the axis of rotation. The cause of the
vibration will be understood after the lecture on centrifugal force
(Lect. XVII.).
 LECTURE V.
THE FORCE OF FRICTION.

The Nature of Friction.—The Mode of

Experimenting.—Friction is proportional to
the pressure.—A more accurate form of
the Law.—The Coefficient varies with the
weights used.—The Angle of Friction.—
Another Law of Friction.—Concluding
Remarks.

THE NATURE OF FRICTION.

112. A discussion of the force of friction is a necessary
preliminary to the study of the mechanical powers which we shall
presently commence. Friction renders the inquiry into the
mechanical powers more difficult than it would be if this force were
absent; but its effects are too important to be overlooked.

Fig. 31.
 113. The nature of friction may be understood by Fig. 31, which
represents a section of the top of a table of wood or any other
substance levelled so that c d is horizontal; on the table rests a block
a of wood or any other substance. To a a cord is attached, which,
after passing over a pulley p, is stretched by a suspended weight b.
If the magnitude of b exceeds a certain limit, then a is pulled along
the table and b descends; but if b be smaller than this limit, both a
and b remain at rest. When b is not heavy enough to produce motion
it is supported by the tension of the cord, which is itself neutralized
by the friction produced by a certain coherence between a and the
table. Friction is by this experiment proved to be a force, because it
prevents the motion of b. Indeed friction is generally manifested as a
force by destroying motion, though sometimes indirectly producing
it.
114. The true source of the force lies in the inevitable roughness
of all known surfaces, no matter how they may have been wrought.
The minute asperities on one surface are detained in corresponding
hollows in the other, and consequently force must be exerted to
make one surface slide upon the other. By care in polishing the
surfaces the amount of friction may be diminished; but it can only be
decreased to a certain limit, beyond which no amount of polishing
seems to produce much difference.
115. The law of friction under different conditions must be
inquired into, in order that we may make allowance when its effect is
of importance. The discussion of the experiments is sometimes a
little difficult, and the truths arrived at are principally numerical, but
we shall find that some interesting laws of nature will appear.

THE MODE OF EXPERIMENTING.

116. Friction is present between every pair of surfaces which are
in contact: there is friction between two pieces of wood, and
between a piece of wood and a piece of iron; and the amount of the
force depends upon the character of both surfaces. We shall only
experiment upon the friction of wood upon wood, as more will be
learned by a careful study of a special case than by a less minute
examination of a number of pairs of different substances.
117. The apparatus used is shown in Fig. 32. A plank of pine
6' × 11" × 2" is planed on its upper surface, levelled by a spirit-level,
and firmly secured to the framework at a height of about 4' from the
ground. On it is a pine slide 9" × 9", the grain of which is crosswise
to that of the plank; upon the slide the load a is placed. A rope is
attached to the slide, which passes over a very freely mounted cast
iron pulley c, 14" diameter, and carries at the other end a hook
weighing one pound, from which weights b can be suspended.
118. The mode of experimenting consists in placing a certain
load upon a, and then ascertaining what weight applied to b will
draw the loaded slide along the plane. As several trials are generally
necessary to determine the power, a rope is attached to the back of
the slide, and passes over the two pulleys d; this makes it easy for
the experimenter, when applying the weights at b, to draw back the
slide to the end of the plane by pulling the ring e: this rope is of
course left quite slack during the process of the experiment, since
the slide must not be retarded. The loads placed upon a during the
series of experiments ranged between one stone and eight stone. In
the loads stated the weight of the slide itself, which was less than 1
lb., is always included. A variety of small weights were provided for
the hook b; they consisted of 0·1, 0·5, 1, 2, 7, and 14 lbs. There is
some friction to be overcome in the pulley c, but as the pulley is
comparatively large its friction is small, though it was always allowed
for.
 Fig. 32.
119. An example of the experiments made is thus described. A
weight of 56 lbs. is placed upon the slide, and it is found on trial that
29 lbs. on b (including the weight of the hook itself) is sufficient to
start the slide; this weight is placed upon the hook pound by pound,
care being taken to make each addition gently.
120. Experiments were made in this way with various weights
upon a, and the results are recorded in Table I.

Table I.—Friction.
Smooth horizontal surface of pine 72" × 11";
slide also of pine 9" × 9"; grain crosswise;
slide is not started; force acting on slide is
gradually increased until motion
commences.
 Load on
Force Force
slide
necessary necessary
Number of in lbs., Mean
to move to move
Experiment. including values.
slide. slide.
weight
1st Series. 2nd Series.
of slide
1 14 5 8 6·5
2 28 15 16 15·5
3 42 20 15 17·5
4 56 29 24 26·5
5 70 33 31 32·0
6 84 43 33 38·0
7 98 42 38 40·0
8 112 50 33 41·5

In the first column a number is given to each experiment for

convenience of reference. In the second column the load on the
slide is stated in lbs. In the third column is found the force necessary
to overcome the friction. The fourth column records a second series
of experiments performed in the same manner as the first series;
while the last column shows the mean of the two frictions.
121. The first remark to be made upon this table is, that the
results do not appear satisfactory or concordant. Thus from 6 and 7
of the 1st series it would appear that the friction of 84 lbs. was 43
lbs., while that of 98 lbs. was 42 lbs., so that here the greater weight
appears to have the less friction, which is evidently contrary to the
whole tenor of the results, as a glance will show. Moreover the
frictions in the 1st and the 2nd series do not agree, being generally
greater in the former than in the latter, the discordance being
especially noticeable in experiment 8, where the results were 50 lbs.
and 33 lbs. In the final column of means these irregularities are
lessened, for this column shows that the friction increases with the
weight, but it is sufficient to observe that as the difference of the 1st
and the 2nd is 9 lbs., and that of the 2nd and the 3rd is only 2 lbs.,
the discovery of any law from these results is hopeless.
122. But is friction so capricious that it is amenable to no better
law than these experiments appear to indicate? We must look a little
more closely into the matter. When two pieces of wood have
remained in contact and at rest for some time, a second force
besides friction resists their separation: the wood is compressible,
the surfaces become closely approximated, and the coherence due
to this cause must be overcome before motion commences. The
initial coherence is uncertain; it depends probably on a multitude of
minute circumstances which it is impossible to estimate, and its
presence has vitiated the results which we have found so
unsatisfactory.
123. We can remove these irregularities by starting the slide at
the commencement. This may be conveniently effected by the screw
shown at f in Fig. 32; a string attached to its end is fastened to the
slide, and by giving the handle of the screw a few turns the slide
begins to creep. A body once set in motion will continue to move
with the same velocity unless acted upon by a force; hence the
weight at b just overcomes the friction when the slide moves
uniformly after receiving a start: this velocity was in one case of
average speed measured to be 16 inches per minute.
124. Indeed in no case can the slide commence to move unless
the force exceed the friction. The amount of this excess is
indeterminate. It is certainly greater between wooden surfaces than
between less compressible surfaces like those of metals or glass. In
the latter case, when the force exceeds the friction by a small
amount, the slide starts off with an excessively slow motion; with
wood the force must exceed the friction by a larger amount before
the slide commences to move, but the motion is then comparatively
rapid.
125. If the power be too small, the load either does not continue
moving after the start, or it stops irregularly. If the power be too
great, the load is drawn with an accelerated velocity. The correct
amount is easily recognised by the uniformity of the movement, and
even when the slide is heavily laden, a few tenths of a pound on the
power hook cause an appreciable difference of velocity.
126. The accuracy with which the friction can be measured may
be appreciated by inspecting Table II.

Table II.—Friction.
Smooth horizontal surface of pine 72" × 11";
slide also of pine 9" × 9"; grain crosswise;
slide started; force applied is sufficient to
maintain uniform motion of the slide.

Load on
Force Force
slide
necessary to necessary to
Number of in lbs., Mean
maintain maintain
Experiment. including values.
motion. motion.
weight
1st Series. 2nd Series.
of slide
1 14 4·9 4·9 4·9
2 28 8·5 8·6 8·5
3 42 12·6 12·4 12·5
4 56 16·3 16·2 16·2
5 70 19·7 20·0 19·8
6 84 23·7 23·0 23·4
7 98 26·5 26·1 26·3
8 112 29·7 29·9 29·8

127. Two series of experiments to determine the power

necessary to maintain the motion have been recorded. Thus, in
experiment 7, the load on the slide being 98 lbs., it was found that
26·5 lbs. was sufficient to sustain the motion, and a second trial
being made independently, the power found was 26·1 lbs.: a mean
of the two values, 26·3 lbs., is adopted as being near the truth. The
greatest difference between the two series, amounting to 0·7 lb., is
found in experiment 6; a third value was therefore obtained for the
friction of 84 lbs.: this amounted to 23·5 lbs., which is intermediate
between the two former results, and 23·4 lbs., a mean of the three,
is adopted as the final result.
128. The close accordance of the experiments in this table shows
that the means of the fifth column are probably very near the true
values of the friction for the corresponding loads upon the slide.
129. The mean frictions must, however, be slightly diminished
before we can assert that they represent only the friction of the
wood upon the wood. The pulley over which the rope passes turns
round its axle with a small amount of friction, which must also be
overcome by the power. The mode of estimating this amount, which
in these experiments never exceeds 0·5 lb., need not now be
discussed. The corrected values of the friction are shown in the third
column of Table III. Thus, for example, the 4·9 lbs. of friction in
experiment 1 consists of 4·7, the true friction of the wood, and 0·2,
which is the friction of the pulley; and 26·3 of experiment 7 is
similarly composed of 25·8 and 0·5. It is the corrected frictions
which will be employed in our subsequent calculations.

FRICTION IS PROPORTIONAL TO
THE PRESSURE.
130. Having ascertained the values of the force of friction for
eight different weights, we proceed to inquire into the laws which
may be founded on our results. It is evident that the friction
increases with the load, of which it is always greater than a fourth,
and less than a third. It is natural to surmise that the friction (F) is
really a constant fraction of the load (R)—in other words, that
F = kR, where k is a constant number.
131. To test this supposition we must try to determine k; it may
be ascertained by dividing any value of F by the corresponding value
of R. If this be done, we shall find that each of the experiments
yields a different quotient; the first gives 0·336, and the last 0·262,
while the other experiments give results between these extreme
values. These numbers are tolerably close together, but there is still
sufficient discrepancy to show that it is not strictly true to assert that
the friction is proportional to the load.
132. But the law that the friction varies proportionally to the
pressure is so approximately true as to be sufficient for most
practical purposes, and the question then arises, which of the
different values of k shall we adopt? By a method which is described
in the Appendix we can determine a value for k which, while it does
not represent any one of the experiments precisely, yet represents
them collectively better than it is possible for any other value to do.
The number thus found is 0·27. It is intermediate between the two
values already stated to be extreme. The character of this result is
determined by an inspection of Table III.
The fourth column of this table has been calculated from the
formula F = 0·27 R. Thus, for example, in experiment 5, the friction
of a load of 70 lbs. is 19·4 lbs., and the product of 70 and 0·27 is
18·9, which is 0·5 lb. less than the true amount. In the last column
of this table the discrepancies between the observed and the
calculated values are recorded, for facility of comparison. It will be
observed that the greatest difference is under 1 lb.

Table III.—Friction.
Friction of pine upon pine; the mean values of
the friction given in Table II. (corrected for
the friction of the pulley) compared with
the formula F = 0·27 R.
 R.
Discrepancies
Total Corrected F.
between the
Number of load mean Calculated
observed and
Experiment. on value of value
calculated
slide in friction. of friction.
frictions.
lbs.
1 14 4·7 3·8 -0·9
2 28 8·2 7·6 -0·6
3 42 12·2 11·3 -0·9
4 56 15·8 15·1 -0·7
5 70 19·4 18·9 -0·5
6 84 23·0 22·7 -0·3
7 98 25·8 26·5 +0·7
8 112 29·3 30·2 +0·9

133. Hence the law F = 0·27 R represents the experiments with

tolerable accuracy; and the numerical ratio O·27 is called the
coefficient of friction. We may apply this law to ascertain the friction
in any case where the load lies between 14 lbs. and 112 lbs.; for
example, if the load be 63 lbs., the friction is 63 × 0·27 = 17·0.
134. The coefficient of friction would have been slightly different
had the grain of the slide been parallel to that of the plank; and it of
course varies with the nature of the surfaces. Experimenters have
given tables of the coefficients of friction of various substances,
wood, stone, metals, &c. The use of these coefficients depends upon
the assumption of the ordinary law of friction, namely, that the
friction is proportional to the pressure: this law is accurate enough
for most purposes, especially when used for loads that lie between
the extreme weights employed in calculating the value of the
coefficient which is employed.

A MORE ACCURATE LAW OF FRICTION.

 135. In making one of our measurements with care, it is unusual
to have an error of more than a few tenths of 1 lb. and it is hardly
possible that any of the mean frictions we have found should be in
error to so great an extent as 0·5 lb. But with the value of the
coefficient of friction which is used in Table III., the discrepancies
amount sometimes to 0·9 lbs. With any other numerical coefficient
than 0·27, the discrepancies would have been even still more
serious. As these are too great to be attributed to errors of
experiment, we have to infer that the law of friction which has been
assumed cannot be strictly true. The signs of the discrepancies
indicate that the law gives frictions which for small loads are too
small, and for large loads are too great.
136. We are therefore led to inquire whether some other relation
between F and R may not represent the experiments with greater
fidelity than the common law of friction. If we diminished the
coefficient by a small amount, and then added a constant quantity to
the product of the coefficient and the load, the effect of this change
would be that for small loads the calculated values would be
increased, while for large loads they would be diminished. This is the
kind of change which we have indicated to be necessary for
reconciliation between the observed and calculated values.
137. We therefore infer that a relation of the form F = x + y R
will probably express a more correct law, provided we can find x and
y. One equation between x and y is obtained by introducing any
value of R with the corresponding value of F, and a second equation
can be found by taking any other similar pair. From these two
equations the values of x and of y may be deduced by elementary
algebra, but the best formula will be obtained by combining together
all the pairs of corresponding values. For this reason the method
described in the Appendix must be used, which, as it is founded on
all the experiments, must give a thoroughly representative result.
The formula thus determined, is

F = 1·44 + 0·252 R.
 This formula is compared with the experiments in Table IV.

Table IV.—Friction.
Friction of pine upon pine; the mean values of
the friction given in Table II. (corrected for
the friction of the pulley) compared with
the formula F = 1·44 + 0·252 R.

R.
Discrepancies
Total Corrected F.
between the
Number of load mean Calculated
observed and
Experiment. on value of value
calculated
slide in friction. of friction.
frictions.
lbs.
1 14 4·7 5·0 +0·3
2 28 8·2 8·5 +0·3
3 42 12·2 12·0 -0·2
4 56 15·8 15·6 -0·2
5 70 19·4 19·1 -0·3
6 84 23·0 22·6 -0·4
7 98 25·8 26·1 +0·3
8 112 29·3 29·7 +0·4

The fourth column contains the calculated values: thus, for

example, in experiment 4, where the load is 56 lbs., the calculated
value is 1·44 + 0·252 × 56 = 15·6; the difference 0·2 between this
and the observed value 15·8 is shown in the last column.
138. It will be noticed that the greatest discrepancy in this
column is 0·4 lbs., and that therefore the formula represents the
experiments with considerable accuracy. It is undoubtedly nearer the
truth than the former law (Art. 132); in fact, the differences are now
such as might really belong to errors unavoidable in making the
experiments.
 139. This formula may be used for calculating the friction for any
load between 14 lbs. and 112 lbs. Thus, if the load be 63 lbs., the
friction is 1·44 + 0·252 × 63 = 17·3 lbs., which does not differ much
from 17·0 lbs., the value found by the more ordinary law. We must,
however, be cautious not to apply this formula to weights which do
not lie between the limits of the greatest and least weight used in
those experiments by which the numerical values in the formula
have been determined; for example, to take an extreme case, if R =
0, the formula would indicate that the friction was 1·44, which is
evidently absurd; here the formula errs in excess, while if the load
were very large it is certain the formula would err in defect.

THE COEFFICIENT VARIES WITH

THE WEIGHTS USED.
140. In a subsequent lecture we shall employ as an inclined
plane the plank we have been examining, and we shall require to
use the knowledge of its friction which we are now acquiring. The
weights which we shall then employ range from 7 lbs. to 56 lbs.
Assuming the ordinary law of friction, we have found that 0·27 is the
best value of its coefficient when the loads range between 14 lbs.
and 112 lbs. Suppose we only consider loads up to 56 lbs., we find
that the coefficient 0·288 will best represent the experiments within
this range, though for 112 lbs. it would give an error of nearly 3 lbs.
The results calculated by the formula F = 0·288 R are shown in
Table V., where the greatest difference is 0·7 lb.

Table V.—Friction.
Friction of pine upon pine; the mean values of
the friction given in Table II. (corrected for
the friction of the pulley) compared with
the formula F = 0·288 R
 R.
Discrepancies
Total Corrected F.
between the
Number of load mean Calculated
observed and
Experiment. on value of value
calculated
slide in friction. of friction.
frictions.
lbs.
1 14 4·7 4·0 -0·7
2 28 8·2 8·1 -0·1
3 42 12·2 12·1 -0·1
4 56 15·8 16·1 +0·3

141. But we can replace the common law of friction by the

more accurate law of Art. 137, and the formula computed so as to
best harmonise the experiments up to 56 lbs., disregarding the
higher loads, is F = 0·9 + 0·266 R. This formula is obtained by the
method referred to in Art. 137. We find that it represents the
experiments better than that used in Table V. Between the limits
named, this formula is also more accurate than that of Table IV. It is
compared with the experiments in Table VI., and it will be noticed
that it represents them with great precision, as no discrepancy
exceeds 0·1.

Table VI.—Friction.
Friction of pine upon pine; the mean values of
the friction given in Table II. (corrected for
the friction of the pulley) compared with
the formula F = 0·9 + 0·266 R.
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

testbankdeal.com