[ Garratt / Marvels and Mysteries of Instinct ]
This volume, the full title of which is Marvels and Mysteris of Instinct: or, Curiosities of Animal Life was first published in 1856 with a second edition in 1857, and a 3rd edition in 1862. The text below is from that 3rd edition (London: Longman, Green, Longman, and Roberts).
I can find no information about this author. Interestingly, the only illustration in this work (aside from a few geometrical diagrams) is the frontispiece, which shows the author (identified only as "the author" in the caption) taking notes as he watches a family of partridges.
The first mention of Newfoundlands comes during a discussion of the role of instinct versus reason in animals:
The Newfoundland dog has a sagacity that is remarkably strong and humane in its character. This animal appears as if designed to be a companion to man, but more particularly when he is exposed to the perils of the water. With semi-webbed feet, which make him a good swimmer, and an inclination to enter the water, this element seems half natural to his nature. It is when persons are in the act of drowning that the sagacity of this dog displays itself most strongly, and innumerable lives has it saved from a watery grave.
One instance will serve our purpose as well as a hundred which might be enumerated. A singular case is given of a person who was travelling in Holland, and accompanied by a Newfoundland dog. Not taking proper heed to his steps in an evening walk along a high bank by the side of one of those deep canals common in the country, his foot slipped, letting him into the deep with a plunge, and being unable to swim, the fish's element soon deprived him of his senses. In the mean time the sagacious animal had no sooner discovered the danger to which his master was exposed, than he was in the water, and engaged in the struggle to rescue him from his peril. A party at a distance saw the faithful servant at one moment pushing, and at another dragging the body towards a small creek, where at length he succeeded in landing his charge, and placing it as far from the water as possible. This being done, the dog first shook himself, and then licked the hands and face of his apparently dead lord. The body being conveyed to a neighbouring house, the efforts to restore the lost senses were successful.
From the marks of teeth on the body, it appeared that the dog had taken his first hold on the shoulder; but, finding that this did not keep the head out of the water, the instinct or the intelligence, — which? — of the animal prompted him to change his grasp from the shoulder to the neck, by which he was enabled to raise the head above the strangling liquid for the distance of about a quarter of a mile.
That this brute creature should have followed his master with such promptness is remarkable enough; but that he should have taken the precaution to keep his head above water, as if he knew that if he did not he would perish, is truly marvellous. What was the motor power in this case? Was it an impulse of instinct, to which the dog rendered a blind obedience? Or was it an act purely of intelligence? Our opinion is that it was a mixed operation, and that both instinct and intelligence had a share in the business. That much of it was owing to instinct, we have the proof in the fact that the Newfoundland dog is not only always found ready to enter the water for such purposes, but even has a strong inclination for it, without either being commanded or directed, and without experience or a precedent. Perhaps the intelligence of this dog is more flexible and accommodating than that of any other species, but yet how infinitely short does it come of the universal! It has its limits, and if these limits are compared with the range man's reason takes, they will appear exceedingly narrow. We do not think, therefore, that the apparent sagacity in the feats of this fine animal is to be taken solely as the result of a purely reasoning faculty. The operations prove too much. (104 - 106)
(The above anecdote first appeared, to the best of my knowledge, in Biographical Sketches and Authentic Anecdotes of Dogs (1829) by Captain Thomas Brown, a Scottish naturalist; this work is discussed here at The Cultured Newf. The anecdote was repeated by a number of other writers during the 19th Century.)
There is one other brief mention of Newfs in Garrat's book:
The pointer possesses a considerable share of intelligence, since he can be trained to obedience; but he, again, cannot be taught to perform any task that does not come within the province of his proper instinct; for instance, this dog, unlike the Newfoundland, will not fetch or carry any object in his mouth, not even the game he has pointed. (114)