cause you think it improbable that they will ever pay dividends. You part from your dearest friend with a smile because you think it extremely improbable that anything will prevent your meeting again on the day appointed. Questions of probability are something more than a mathematician's pastime.

Many questions of fact, past and present, far and near, have not yet been settled, and may never be settled beyond a certain degree of probability. But there is another class of questions which we do not hope to settle beyond a degree of probability. Not because they do not involve facts, but because we recognize that the facts are beyond our reach, or because we know that the future alone will determine them, while our interest in them is purely a present one. For instance, we are content for the present to speculate upon the probable internal structure of the earth. Perhaps some day a serious attempt will be made to arrive at the facts. Again, we are confronted with the question of what the weather will be to-morrow. Now, it will either rain or not rain, but we cannot wait to learn the fact; and we may not be half so much interested in knowing the fact when it comes as we are now in knowing the probability, for now only can we decide the question whether we shall go on our journey provided with an umbrella or not. Governing our present action by the probability we make up our minds to accept the future fact with as little concern as possible.

How do we determine the probability, or, as we often say, the chances, that a thing is thus and thus or that an event will happen in a certain manner? By observation and experience, by induction and deduction.

Every imperfect induction is merely the expression of a probability. Every deduction carried beyond the range of actual experience is likewise only a probability. There is another phase of this matter. There is a principle of reasoning, how obtained we cannot discuss here, which declares that "we must treat equals equally, and what we know of one case may be affirmed of every case resembling it in the necessary circumstances." Of course experiment may be necessary to determine whether things are equal or not, but starting with this principle we calculate probabilities without experimentation. Indeed in many cases the experiment proves nothing whatever in regard to future results, it only proves the principle. I toss a penny into the air. It has two sides and so far as I know they are equal. I know it will fall upon one side or the other. The other conditions I do not know and can not control, and so I say that there is only an even chance that the head will fall uppermost. Suppose it falls so. I conclude nothing whatever from that in regard to the manner in which it will fall a second time. Suppose I toss it up ten times and the head comes up five times, the tail five times, can I reason that it will be so the next ten times? Not at all. Not at all. I know, each time I toss it, that there is an even chance of the head coming uppermost. Therefore it is entirely possible that it will come uppermost ten times in succession. But because the chances are even I say that such a result, though possible, is improbable; that it is most probable that head and tail will each come uppermost five times; that the next greatest probability is that one will come uppermost six times and the other four; that it is most

improbable that either one will come up ten times in succession. By such laws of mere probability, without any degree of certainty whatever, are we compelled to determine a thousand acts of our everyday life. Though often a matter of mathematical computation, serious errors have been made and there is room for argument even here.

There is still more room for argument in cases that are not susceptible of mathematical demonstration. Take a prophecy, as for example that the world will come to an end next week or in the year 2000, or let some member of the class write a prophecy, and then debate upon the probability of its being fulfilled. Or take any current newspaper report that is of a surprising or sensational nature and argue from antecedent probabilities that it is or is not true. Argumentative exercises of this nature may be made extremely interesting and instructive.

"Conduct is three-fourths of life," says Matthew Arnold. Another amends this and says, "Conduct is the whole of life." Living means something more than being; it means something more than knowing or believing it means action, conduct, behavior. The man who knows without acting upon that knowledge is as censurable as the man who acts without knowledge. And what does the Apostle James say of faith without works?

The office of persuasive discourse is to arouse men to action. Exposition, we said, presupposes some degree of ignorance on the part of those addressed, and argumentation presupposes error. Persuasion presupposes indifference, inaction, or misdirected action; it appeals to the emotions, the feelings. Strange as it seems, we may know a truth, we may firmly believe it to be truth, and yet fail to take it home to ourselves, to act upon it,